Ending Homelessness in America: An Evidence Framework for Communities, Funders, and Policymakers

By Common Ladder · May 26, 2026 · 45 min read

The anchor document of the Common Ladder publishing portfolio on ending homelessness. A synthesis for senior foundation program officers, HUD and state housing agency officials, CoC executive directors with policy ambitions, and the legislative staff and philanthropic boards who will decide whether the next decade looks like the last one.

Thesis

Homelessness in the United States is solvable. The evidence on what works exists, has existed for two decades, and is more rigorous now than it has ever been. What is missing is not knowledge. What is missing is the political economy — the funder discipline, the policy architecture, the operational fidelity, the workforce investment, and the structural housing supply — to apply what we already know at the scale the problem demands.

This document is the case that says so, in full. It is written for the people who can change what gets funded, what gets built, and what gets measured: senior foundation program officers, senior HUD and state housing agency officials, CoC executive directors with policy ambitions, the legislative staff who will draft the next reauthorization, and the philanthropic boards that will decide whether the next decade looks like the last one.

Executive Summary

Part I — The Architecture of the Problem

1.1 What Homelessness Is — And Isn't

Homelessness in the United States is the predictable output of a set of structural conditions: a housing market in which rents have outpaced the incomes of lower-decile households for two decades, an income floor that has not kept pace with the cost of stable housing in most major markets, structural racism that compounds housing exclusion across generations, healthcare gaps that leave the most vulnerable populations untreated until they are in crisis, and safety net architecture that fails to catch people whose income is interrupted by job loss, illness, or family rupture.^{[f1][f9][f16]} The evidence supporting this systemic framing is no longer marginal. It is the dominant view in the peer-reviewed housing-policy literature.

This matters because the alternative framing — that homelessness is primarily caused by individual choices, behavioral health conditions, or moral failings — produces a different and ineffective set of responses. Treatment-first models, sobriety contingencies, work requirements, and enforcement-led encampment clearings are responses to a problem that the evidence does not support is present in the form the responses assume. Housing First — the framework that prioritizes immediate access to permanent housing without preconditions — has been tested in multiple randomized controlled trials over two decades and produces dramatically better housing outcomes than treatment-first models.^[f2][f3] The treatment-first frame is not just less effective. It is a tangible mismatch between the response and the problem.

This does not mean individuals do not have behavioral health needs. Many people experiencing homelessness do, and the evidence on Permanent Supportive Housing with intensive case management is specifically the evidence that addresses this. It means that the access pathway to housing is the wrong place to put a treatment gate. Housing is the platform from which treatment becomes possible. The order matters.^{[f2][services]}

1.2 The System Is a Flow, Not a Stock

Homelessness at a community level is a flow problem. At any moment there is a number of people experiencing homelessness — the stock. That stock changes based on three forces: inflow (new entries into homelessness), exits (placements into permanent housing), and prevention (averted entries). A community reaches what the field calls functional zero when exits and prevention combined exceed inflow — meaning that whatever number of people become homeless in a given month, the system has the placement capacity to house at least that many, and the people who enter homelessness do so briefly and do not return.

This reframing matters more than it appears to. The instinct in a homelessness crisis — particularly a politically visible one — is to focus on the stock: how many tents are visible, how many shelter beds are full, how many people did the Point-in-Time count find. These are real and useful measures, but they tell almost nothing about whether the system is improving. A community that houses one thousand people in a year while two thousand new people become homeless has a stock that grew. A community that houses two hundred people in a year while one hundred new people become homeless has a stock that shrank. The first community looks busier and is failing. The second community looks quieter and is winning.

The strategic implication is that every community has three levers — reduce inflow, increase exits, accelerate throughput — and the right combination depends on local conditions. In a community where homelessness is increasing rapidly from a low base, prevention and diversion offer the highest leverage. In a community with entrenched chronic homelessness, exits to PSH are the binding constraint. In a community whose by-name list and coordinated entry already work, throughput improvements — clearing bottlenecks at the housing match stage, accelerating lease-up — can move the count more than any new intervention.

1.3 The 2024 Baseline

The federal 2024 Annual Homelessness Assessment Report recorded the highest single-night count of people experiencing homelessness since national reporting began in 2007: approximately 770,000 people on a January night, an 18 percent increase over the prior year.^[f1] The composition matters as much as the total.

Family homelessness rose 39 percent in a single year — the steepest increase of any population category, and a reversal of a decade of progress.^[f1][f8] Children under 18 in homeless families rose 33 percent. Unaccompanied youth reached the highest recorded count, though the Point-in-Time methodology is known to undercount youth severely; survey-based prevalence estimates suggest the true number is many times higher than the PIT captures.^[f11] Veterans homelessness, alone among the major subpopulations, declined: down 7.6 percent overall and down 10.7 percent among unsheltered veterans.^[f10]

The veterans trend is the proof of concept that ending homelessness for a defined population is possible in the United States. It was achieved through a specific combination: dedicated vouchers (HUD-VASH), an integrated services system (the VA), real-time by-name data, an explicit functional zero target, and named public accountability for the outcome.^[f10] The combination is replicable. The political constituency that produced it — bipartisan support for housing veterans — is not automatically replicable for non-veteran populations, which is part of why the veterans model has not extended to families, youth, or chronic populations at the same scale.

Two facts about 2024 should sit alongside the count. First, the 2025 Annual Homelessness Assessment Report has not been released as of this writing. Local Point-in-Time counts were conducted in January 2025; the national synthesis has not been published. This is a data governance problem that compounds every other problem — the field is operating critical funding and policy decisions without national outcome data. Second, the recorded count is itself a substantial undercount. McKinney-Vento school identifications captured approximately 1.5 million children and youth experiencing homelessness in the 2022–23 school year — roughly eight times the PIT count for that year.^[f11] The actual scale of housing instability is larger than the PIT reflects, and the policy response should be sized to the actual scale.

1.4 What "Ending Homelessness" Means Precisely

The phrase "ending homelessness" deserves precision. It does not mean that no one ever loses housing. It means that, at a community level, homelessness has become rare, brief, and non-recurring — what the field calls functional zero.^{[functional_zero]} Rare means the inflow is small relative to community population. Brief means people who enter homelessness move to housing in weeks, not years. Non-recurring means people who exit homelessness do not return.

This precision matters because the alternative framing — "homelessness will always exist; there's no point in trying to end it" — is rhetorically sticky and policy-toxic. It is also wrong on the evidence. Specific communities have achieved functional zero for specific populations: veterans in 12 or more communities, chronic homelessness in several.^[f10] No community has yet achieved overall functional zero across all populations, and the political and resource conditions to do so have not aligned. But the framework is validated. What remains is execution at scale.

External communications about this work should use the precise framing. "Making homelessness rare, brief, and non-recurring" is more accurate and more defensible than "ending homelessness." Internally, "ending homelessness" remains useful as a moral and motivational frame — it names what the work is for, even when the daily task is incremental.

Part II — What the Evidence Says Works

The homelessness literature has produced a small set of canonical findings that should anchor every funding, program, and policy decision. They are summarized here with their evidence basis and their operational implications.

2.1 Housing First

Housing First is not a program. It is a framework — a set of principles that says the path to recovery, stability, and treatment access runs through housing first, not after. People are not required to be sober, in treatment, or "ready" to receive housing. Lease violations are grounds for termination; behavioral health symptoms and substance use are not. Coordinated entry prioritizes by vulnerability, not by treatment compliance.

The evidence is canonical. The At Home/Chez Soi five-city Canadian RCT, with 2,148 participants and a seven-year follow-up, found that Housing First participants spent 73 percent of their time in stable housing compared with 32 percent in treatment-as-usual — a 41 percentage-point improvement that persisted for years.^[f2] After two years, 62 percent of Housing First participants were continuously housed, against 31 percent of controls. The Santa Clara County PSH RCT, with a different population and study design, found 86 percent of intervention participants were ever housed during the four-year trial, compared with 36 percent of controls; intervention participants remained stably housed for 84 percent of any given one-year period.^[f2][f6]

The evidence on non-housing outcomes is more nuanced. Systematic reviews and meta-analyses of Housing First RCTs find consistent large effects on housing stability but inconsistent effects on mental health, substance use, and self-rated quality of life when compared with treatment-as-usual controls.^[f3] This finding is sometimes used to suggest Housing First "doesn't work" — a mischaracterization. Housing First reliably delivers what it is designed to deliver: housing. It does not reliably deliver behavioral health recovery on its own. That outcome requires co-occurring treatment engagement, which Housing First makes accessible but does not guarantee. The honest framing for funders and policymakers: Housing First is the housing intervention. Behavioral health interventions are separate, complementary, and required for the highest-acuity populations — see Section 2.6.

What Housing First also reliably reduces is non-routine healthcare utilization. The At Home/Chez Soi seven-year follow-up found high-need Housing First with ACT participants had rate ratios of 0.66 for primary care visits and 0.64 for specialist visits compared with TAU — the kind of reduction that converts chaotic crisis-driven healthcare into appropriate ongoing care.^[f2][f3]

The operational implication: Housing First should be the standard implementation framework for every CoC, shelter, and housing program in the United States. Deviations should require justification grounded in evidence — not in administrative preference, not in moral judgment about who deserves housing, not in political pressure to look tough on homelessness. The evidence for Housing First is among the strongest in the social policy literature. The post-2024 federal political environment is creating pressure to deviate from it in the direction of treatment-first and sobriety-contingent models. Communities that maintain Housing First fidelity under that pressure will produce better outcomes than those that don't.

2.2 Permanent Supportive Housing for Chronically Homeless Adults

For adults with histories of chronic homelessness — meaning continuous or repeated episodes of homelessness over at least a year, typically accompanied by a serious mental illness, substance use disorder, or chronic medical condition — Permanent Supportive Housing combines a permanent housing subsidy with onsite or wraparound supportive services. The evidence base is canonical: the Lancet Public Health systematic review reported rate ratios of 1.13 for moderate-need and 1.42 for high-need participants compared with usual care on long-term housing stability.^[f6] The National Academies of Sciences, Engineering, and Medicine PSH evidence review reached the same directional conclusion.^[f6]

The cost-offset evidence is also canonical, with the caveat that it must be reported precisely. Culhane, Metraux, and Hadley's 2002 analysis of New York City's NY/NY supportive housing program — the foundational quasi-experimental study — tracked 4,679 placements across eight linked administrative databases and found regression-adjusted reductions of $12,146 per placement per year across shelter, healthcare, mental health, and corrections costs, against annual housing costs of approximately $17,277.^[f4] Adjusting for housing turnover, the net offset reached approximately 94 percent of housing costs within two years — meaning the net public cost of a PSH placement, after service-cost reductions, was approximately $995 per unit per year. The Community Guide systematic economic review of twenty PSH studies in 2022 found median program cost of $16,479 per person per year against median economic benefit of $18,247 — net positive, consistent with the NYC finding.^[f4][f5]

Three precisions are essential when citing this evidence. First, the savings are not in the same budget as the cost. They accrue to Medicaid, state mental health, corrections, and to a smaller extent shelter — not to the housing budget that funds the unit. This is the cross-agency funding problem, and it is the primary structural barrier preventing cost-offset evidence from translating into investment.^[f17] Second, the Medicaid outpatient costs increased in the NYC data — by 76 percent — because newly housed people accessed appropriate outpatient care that they previously could not. This is a positive finding, not a negative one, but it must be disclosed when citing the net savings number. The right framing is reduction in crisis utilization and appropriate increase in outpatient care, not reduction in total healthcare. Third, the foundational evidence is from a specific NYC program in the 1990s, in 1999 dollars. The directional finding generalizes; the specific dollar figures must be understood as a starting point for local cost analysis, not a universal benchmark.

The Permanent Supportive Housing pipeline is the single most important capital investment a community can make for its chronically homeless population. As of 2024, the United States had approximately 412,623 PSH beds, of which roughly 170,000 rely on Continuum of Care program funding.^[f6] The estimated chronically homeless population in 2024 was 152,585 — meaning the existing inventory, in aggregate, approximately matches need, but is geographically maldistributed and underdeveloped in many high-need markets.^[f1][f6] A community-level analysis of "do we have enough PSH" must be done at the community level, not the national.

2.3 Housing Choice Vouchers for Families

For families experiencing homelessness, the evidence is unambiguous: permanent rental subsidy outperforms every alternative. The HUD Family Options Study — a multi-site randomized controlled trial across 12 communities with 2,282 families — found that families randomized to a housing voucher experienced substantially better housing stability than those randomized to community-based rapid rehousing, transitional housing, or usual care (emergency shelter).^[f8] The voucher effects extended beyond housing into children's behavioral and educational outcomes, observed at both 20-month and 37-month follow-ups.

This evidence is not new. The Family Options Study was published in 2016 and updated in 2022 and 2024. It has not changed federal voucher policy. Family homelessness rose 39 percent in 2024 — driven in part by the simple arithmetic that the Housing Choice Voucher program reaches only about one in four households eligible for it.^[f1][f8] When a family on the voucher waiting list — typically a wait of years in major markets — experiences a housing crisis, the system has produced a predictable outcome before the crisis began. The voucher was the answer. The voucher was not available. The family entered shelter.

The operational implication for communities is that voucher access strategy matters as much as program design. CoCs that build active relationships with their public housing authorities, that secure project-based voucher commitments for new development, that supplement federal vouchers with state or local rental assistance, and that advocate for emergency housing voucher expansion produce better family outcomes than CoCs that treat the voucher pool as fixed. The national policy implication is that voucher expansion is the highest-leverage investment available for family homelessness. The Housing Choice Voucher program is rationed by appropriation. The rationing is the policy choice that produces 39 percent family homelessness increases.

2.4 Targeted Prevention with Emergency Financial Assistance

The most underappreciated finding of the last decade is that homelessness prevention works at a scale and effect size that rivals — and in cost-effectiveness, exceeds — most exit-side interventions. The evidence comes from two converging studies: a 2016 Science paper by Evans, Sullivan, and Wallskog using a natural experiment at the Chicago Homelessness Prevention Call Center, and a 2025 Review of Economics and Statistics paper by Phillips and Sullivan reporting a randomized controlled trial in Santa Clara County, California.^[f22][f23]

The Evans 2016 study exploited a quasi-experimental discontinuity: when Chicago prevention funds were exhausted, otherwise-identical callers did not receive assistance. Comparing funded versus unfunded callers, with a sample of 4,448, receiving prevention assistance reduced shelter entry by 76 percent in the six months following the call. The average grant was approximately $800 to $1,000. The estimated cost per homelessness case prevented was approximately $10,500.^[f23]

The Phillips and Sullivan 2025 RCT randomized households at imminent risk of homelessness to receive financial assistance (averaging approximately $2,000, supported by case management) or to a no-assistance control. The trial found an 81 percent reduction in homelessness at six months and a 73 percent reduction at twelve months. The effects persisted beyond the grant period. The authors calculate a Marginal Value of Public Funds of 2.47 — meaning each dollar spent on this intervention returns $2.47 in social benefit, primarily through avoided shelter and emergency service costs.^[f23]

The convergence is significant. Two independent studies, different methodologies, different populations (single adults in Chicago; families in Santa Clara), produced effect sizes of the same order of magnitude — 70 to 80 percent reductions. This kind of convergence is unusual in homelessness research and elevates the policy confidence the findings should command.

The accompanying finding — equally important — is that targeting is inseparable from effect size. Both studies depended on reaching genuinely imminent-risk households. The Shinn-Greer 2013 AJPH study established that statistical risk-screening models substantially outperform unaided worker judgment in predicting which households will enter shelter.^[f22] The Los Angeles County Prevention Targeting Tool, externally evaluated, identifies the top-decile-risk population at roughly five times the homelessness rate of the general low-income population.^[f22] Prevention dollars distributed to households without targeting will produce smaller effects than prevention dollars distributed with statistical targeting, because the counterfactual matters: the dollars are only preventing homelessness if the recipient would otherwise have entered shelter. Targeting plus intervention is the formula. Either alone is incomplete.

The funder implication is large. Targeted emergency financial assistance with statistical screening is among the most cost-effective interventions in the homelessness evidence base. It is also among the most under-resourced. The federal Emergency Rental Assistance program at $46 billion across 12.3 million payments during the pandemic period was the largest prevention investment in U.S. history and demonstrably blunted the homelessness increase that COVID-19 would otherwise have produced.^[f13] No comparable steady-state federal prevention program exists. State and local prevention funding is patchy, often pegged to crisis appropriations, and rarely paired with statistical targeting infrastructure. This is fixable. The evidence is in.

2.5 Diversion at the Front Door

Diversion is a housing-focused problem-solving conversation at the point of shelter entry, designed to help a household resolve the immediate crisis without entering the shelter system at all. It is not a denial of shelter access — that is the misframing that diversion sometimes encounters in advocacy critique — but an alternative path offered at the moment when a household first presents at the system's front door.

The evidence is promising and pre-RCT. A 2023 evaluation by Building Changes and Clarus Research, with 13,876 Washington State families, found that approximately half of families found safe housing quickly through diversion at a median time of 37 days and an average cost of $1,668 per family housed — substantially shorter and cheaper than shelter-based pathways.^[f18] A Notre Dame / J-PAL randomized controlled trial of shelter diversion began in 2024 with results expected in 2026–2027; until then, diversion is best characterized as a strong-promising intervention with limited rigorous causal evidence.^[f18]

The operational implication is to build diversion into coordinated entry intake as a routine first-stage conversation, not a screen-out mechanism. Done well, diversion produces faster housing stabilization for the households for whom a shelter stay would not have been necessary in the first place. Done poorly, it reduces shelter access for households who actually needed it. The line between the two is implementation fidelity — adequate staff training, supportive financial assistance available where it would solve the problem, and unambiguous escalation paths into shelter when diversion is not the right answer.

2.6 Permanent Supportive Housing and the Service Layer

Permanent Supportive Housing is housing plus services. The services are not optional. The original Housing First evidence base — particularly the At Home/Chez Soi trial — evaluated housing combined with Assertive Community Treatment (for high-need participants) or Intensive Case Management (for moderate-need). Housing alone is not what was tested for the chronic and high-acuity population. The "supportive" in PSH is the difference between PSH and a deeply subsidized apartment without services — and the difference matters for outcomes.^{[f2][services]}

The evidence-based service continuum for housing programs runs from low-intensity to high-intensity:

• Housing Navigation for lower-acuity exits, prevention, and rapid rehousing
• Standard Case Management for stable PSH residents and moderate-need rapid rehousing
• Intensive Case Management (ICM) for newly housed residents and moderate-high acuity populations
• Critical Time Intervention (CTI) for system transition moments — discharge from jail, hospital, or shelter into housing
• Assertive Community Treatment (ACT/FACT) for the highest-need, most medically and behaviorally complex residents — typically a multidisciplinary team with low caseload ratios and 24/7 availability

CTI in particular has been evaluated in multiple randomized controlled trials and is now classified as canonical for the population it serves: people transitioning from institutional settings into community housing, where the risk of return to homelessness is concentrated in the first weeks and months.^[services]

The structural problem is that services are chronically underfunded relative to what the evidence-based model requires. A PSH development with underfunded services is not delivering PSH — it is delivering subsidized housing with inadequate support, which produces worse outcomes than the model evidence predicts. The capital-services split — capital funded through one stream (LIHTC, HOME, state housing trust funds), services funded through another (CoC services component, Medicaid, state mental health, philanthropy) — is the operational manifestation of the cross-agency funding problem.^[f17] Communities that solve it require services commitments before approving capital allocations, or use Medicaid to fund services in completed developments, produce more consistent outcomes than communities that fund the building and hope for the services later.

2.7 Rapid Rehousing: Uses and Limits

Rapid Rehousing (RRH) provides time-limited rental assistance and supportive services to move households out of homelessness quickly. The evidence for RRH is mixed and population-dependent. The HUD Family Options Study found that approximately 70 percent of families assigned to RRH exited to permanent housing, with 10 percent returning to homelessness within 12 months — outcomes better than shelter or transitional housing arms, but worse than the voucher arm.^[f7][f8] However, 76 percent of families had moved at least once 12 months after exit; only 24 percent remained in the same unit. Housing instability persists even where technical homelessness does not.

The system-wide return-to-homelessness rate after RRH varies from 10 to 50 percent depending on population, follow-up period, and definition.^[f7] The most common explanation for the range is acuity mismatch. RRH works for lower-acuity households whose homelessness is driven by an income or housing shock that time-limited assistance can resolve. RRH fails for high-acuity households whose needs require permanent subsidy and intensive services. A community that routes high-acuity individuals to RRH because PSH is full produces predictable returns to homelessness, and the failure is not an RRH failure — it is a coordinated-entry failure.

The operational implication is that RRH is a powerful tool when matched to the right population. The matching, not the program, is the leverage point.

2.8 Encampment Policy: Enforcement-Only Doesn't Work

Encampment clearings without simultaneous adequate housing offers cause displacement, not resolution. RAND Corporation's longitudinal data from three Los Angeles neighborhoods documented that encampment clearings produced temporary drops in encampment density lasting two to three months, after which populations returned to nearby areas.^[f14] Where reduced unsheltered counts were achieved — RAND's 2024 data showed a 15 percent year-over-year drop in one corridor — the researchers attributed the change to increased housing placement throughput, not enforcement.

Denver's AIMHigh initiative paired outreach with active housing placement (the House1000 program) and produced 98 percent fewer encampments of 20 or more people and 89 percent fewer encampments of 10 to 20 people — but only because the resolution offered concrete placement, not because the resolution involved enforcement.^[f14] The qualitative finding from interviews with residents of cleared encampments is consistent: residents express strong preference for transitional housing over emergency shelter and cite fear of shelter conditions as a primary reason for remaining unsheltered.

The policy implication after the 2024 Grants Pass decision — which removed the Eighth Amendment limit on criminalizing public camping — is that communities will face increased political pressure to clear encampments. Communities that do so without housing offers will produce displacement that returns within months. Communities that pair resolution with concrete placement, real-time tracking of where people went, and ninety-day confirmation of placement stability will produce different outcomes. The framework should be: no encampment clearing is counted as a system success until placement is confirmed at 90 days.

Part III — The Operational Infrastructure

The evidence on what works at the program level is necessary but not sufficient. The system layer that connects programs to outcomes is where most communities fail. This section names what the operational infrastructure must include for a community to convert programs into system performance.

3.1 Coordinated Entry as System Router

Coordinated Entry (CE) is the federal requirement and the operational mechanism through which a community routes people from the front door of the homelessness system to the right intervention. A high-performing CE produces three things: it identifies every person in need of housing assistance, it assesses their acuity using a consistent validated tool, and it routes them to the appropriate intervention based on acuity and resource availability — not on which provider has the shortest waitlist.

Most CEs do the first imperfectly, the second inconsistently, and the third badly. The most common failure is assignment by availability: when the right intervention for a high-acuity person is PSH and PSH is full, the person gets RRH because RRH has a slot. This produces predictable returns to homelessness, wasted RRH resources on the wrong population, and a deepening of chronicity for the person who needed PSH and didn't get it.^[f7]

A high-performing CE has the following characteristics:

• A validated acuity assessment applied consistently to every household presenting at the system's front door
• A real-time prioritization list, not a waitlist — the highest-priority person gets the next available appropriate resource, regardless of when they entered the system
• Racial equity audited in prioritization decisions, with disparities surfaced and addressed at governance level
• A by-name list integrated with the prioritization process, so every known homeless individual in the community is visible to the coordinator
• Diversion offered as a routine first-stage conversation at intake, not as a back-door screen-out
• Clear escalation pathways when the right intervention is not available — including documented unmet need that drives capital and operating advocacy

CE is also where the most consequential equity decisions in the system get made. The acuity tools commonly used — VI-SPDAT and its successors — have been criticized for racial bias in scoring, and the field is in the middle of a transition to new tools that better correct for these biases. Until that transition is complete, CoCs should be auditing their own prioritization data for racial disparities at every stage: assessment scores, referral rates, time-to-housing, intervention type.

3.2 By-Name Lists as Operational Backbone

The by-name list is the system's operational picture of itself. In a community that has reached or approached functional zero for any population, the by-name list is the prerequisite that made it possible. The list names every individual experiencing homelessness, tracks their current status and time in the system, links them to an assigned navigator, and is updated in real time — not monthly, not at the next case conferencing meeting.^[f12]

The evidence that by-name lists drive outcomes is technically provisional — most of the documentation is from Community Solutions and Built for Zero, both of which have vendor conflicts of interest as the originators of the methodology. The mechanism, however, is operationally obvious: a community that does not know how many people are experiencing homelessness by name cannot manage the system at the individual level required to reach functional zero. The veterans success stories — Hennepin County, Douglas County, the more than twelve communities that have documented functional zero for veterans — all built by-name infrastructure first.^[f10]

The data infrastructure required is modest but the operational discipline is not. By-name lists succeed when there is a single accountable owner of the list, regular case conferencing that uses the list to identify and clear individual cases, and integration with HMIS so the list isn't a parallel spreadsheet but the same data the federal government sees in performance reporting. They fail when the list is treated as a reporting artifact rather than an operational tool, or when it is owned by a vendor or initiative rather than by the CoC.

3.3 HMIS, Data Quality, and System Performance Measures

HMIS is the federally mandated Homeless Management Information System through which CoCs report to HUD. It is also the data infrastructure on which the System Performance Measures (SPMs) — the metrics that gate competitive grant renewal — are calculated. High-performing CoCs treat HMIS as a program management tool, not a compliance burden. They invest in data quality continuously, not seasonally; they disaggregate every key metric by race and ethnicity as a routine governance practice; they track SPMs in real time rather than discovering them at annual reporting; they hold programs accountable to outcomes captured in HMIS.

HMIS data quality is a workforce and management problem more than a technical one. The system is well-defined; the workflows are documented; the federal guidance is robust. What goes wrong is at the data entry layer: undertrained staff, conflicting workflow priorities, missing fields, late entries, inconsistent application of program enrollment definitions. A CoC that wants to know its true return-to-homelessness rate cannot know it from HMIS data that has 30 percent missing fields and three-month entry lags. The fix is operational discipline, supported by training and supervision, not new software.

Recent peer-reviewed analyses of CoC performance have surfaced an important finding for governance: CoC SPM performance is predicted at least as strongly by governance composition and external housing market conditions as by internal program quality.^[f21] Boards with higher non-government stakeholder representation — people with lived experience, providers, faith-based organizations, advocacy groups — show better outcomes than government-dominated boards. Smaller, more focused boards outperform larger ones, after coordination costs exceed the value of added perspectives. Federal funding intensity (CoC dollars per capita homeless) is positively associated with SPM-1 (reduced length of time homeless) and SPM-7 (increased successful placements), with the effect attenuating in high-rent markets where housing supply constraints overwhelm service-side investment.

The implication for CoC leadership: governance design is not a peripheral matter. It is a structural determinant of outcomes. Boards that fail to include people with lived experience, that exclude advocacy and provider perspectives, or that become too large to make decisions are not just suboptimal — they are statistically associated with worse community outcomes.

3.4 Workforce as Service Quality Multiplier

All of the above — coordinated entry, by-name lists, HMIS data quality, intensive case management, ACT teams, PSH services — is delivered by a workforce that is systematically undercompensated relative to the complexity of the work it performs. National Alliance to End Homelessness 2023 survey data (n=5,044 across 1,154 agencies) reported 71 percent of agencies experiencing turnover challenges and 74 percent reporting they were understaffed.^{[f21][workforce]} Annual turnover of 30 to 50 percent is common in many settings. Relationship rupture is itself a driver of housing instability — clients lose continuity with the staff who know their story, and the next staff member starts the relationship-building from zero.

The evidence on staff caseload ratios is clear at the program level: ACT teams require caseloads of approximately 10:1; ICM caseloads should not exceed 20:1; standard case management ratios are typically 35:1. Caseloads that exceed these benchmarks predictably degrade service quality. Licensed clinical roles — psychiatry, therapy — are the scarcest and most under-compensated relative to peer sectors. Secondary traumatic stress and burnout are occupational hazards that organizational supports — supervision, debriefing, mental health benefits, manageable caseloads — meaningfully mitigate.

A community strategy that does not include workforce investment is a strategy that produces excellent programs on paper and degraded programs in practice. High-performing agencies pay above sector median, invest in supervision and clinical consultation, build career pathways, and recruit from the communities they serve. Workforce diversity matters: data on the homelessness workforce is incomplete, but available evidence suggests representation gaps in clinical and leadership roles relative to the populations served.^[workforce]

3.5 Cross-System Cost Accounting

The single most consequential operational capacity a community can build for advocacy purposes is the ability to track public service costs across systems for a defined client population before and after housing placement. The Culhane 2002 NYC study used linked administrative databases across eight systems to produce its findings.^[f4] Almost no CoC has equivalent capability today, even in aggregate, even for its own highest-utilizer cohort. As a result, when a county executive asks "what are these chronically homeless residents costing the system?" the honest answer is usually "we don't know precisely, but national research suggests..." That answer doesn't move budgets.

The data infrastructure required is real but not insurmountable. Shelter data exists in HMIS. Emergency department and hospital data exists in healthcare partner records, sometimes accessible through Medicaid claims if the CoC has data-sharing agreements with the state. Jail and corrections data exists in county records. Mental health utilization exists in state behavioral health data. The work is governance and data-sharing agreements, plus the analytical capacity to aggregate. The reward is a credible cross-agency ROI conversation with the budget holders whose savings PSH actually produces.^[f17]

Part IV — The Structural Conditions

The system layer described in Part III matters because it is the operational machinery through which interventions reach outcomes. But operational excellence cannot end homelessness alone. The structural conditions — housing supply, income adequacy, racial inequality, healthcare access, safety net architecture — produce inflows the system layer must absorb. A community that builds the best possible operational layer on top of broken structural conditions will run a more efficient system that still produces homelessness at scale, because the inflow exceeds anything the system can absorb. This section names what those structural conditions are.

4.1 Housing Supply and Affordability

The most rigorously established structural cause of homelessness at the community level is the relationship between rent and income. The Glynn, Byrne, and Culhane analysis published in Annals of Applied Statistics in 2021 — a Bayesian nonparametric regression on 386 U.S. CoCs using HUD PIT data — found a threshold effect: once median rent exceeds approximately 30 percent of area median income, marginal rent increases produce disproportionate homelessness increases.^[f16] Areas above this threshold contain 15 percent of the U.S. population but 47 percent of people experiencing homelessness — a 3x-plus overrepresentation.

The mechanism is straightforward. As housing costs consume more of a household's income, the financial margin available to absorb shocks shrinks toward zero. When the margin is zero, any shock — job loss, medical emergency, relationship rupture — produces housing loss. The structural question is not whether a person had an individual crisis but whether the housing market left them any margin to survive it. The 30 percent threshold is consistent across the analysis methods used; specific local thresholds may differ but the directional finding is robust.

The supply side of the problem is the United States is short by approximately 7.3 million rental units affordable and available to extremely low-income households at or below 30 percent of area median income. For every 100 such households, roughly 37 affordable and available units exist nationally. LIHTC, the country's largest affordable housing production tool, generates roughly 100,000 units per year nationally, primarily targeting 50 to 60 percent AMI — not the 10 to 30 percent range where homelessness risk concentrates. Public housing has lost an estimated 250,000-plus units since 1990 to demolition, deterioration, and conversion. Naturally occurring affordable housing — older, lower-quality market-rate units — has been lost to gentrification, redevelopment, and conversion at rates that exceed new affordable production.

The policy distinction between "affordable" and "deeply affordable" is essential and frequently lost. Affordable housing in the LIHTC sense — 50 to 60 percent AMI — does not reach the population that becomes homeless. Deeply affordable housing — accessible at 30 percent AMI or lower — requires layered subsidy that the current financing system has never produced at sufficient scale. Advocacy for "more affordable housing" without specifying deep affordability produces units the homeless population cannot afford to enter. Advocacy for ending homelessness must specify deep affordability, with project-based vouchers, operating subsidies, or in-kind construction subsidy layered to reach 30 percent AMI or below.

4.2 Income Poverty and the Low-Wage Economy

The corollary to housing supply is income. A household whose income exceeds 30 percent of area median rent can absorb shocks; a household whose income is below it cannot. SSI benefits — the income source for many people with disabilities experiencing or at risk of homelessness — have not kept pace with housing costs in any major U.S. market. The full SSI benefit is below fair-market rent for a one-bedroom apartment in essentially every metropolitan area. TANF benefits have lost real value over decades; the typical state's TANF benefit is now a fraction of what it was at the program's creation. The federal minimum wage has not increased since 2009.

The implication is that prevention and exit-side interventions interact with income policy. Rapid Rehousing works best when the income side improves during the assistance period; absent income improvement, time-limited assistance produces predictable returns to homelessness at termination. Voucher programs require landlord willingness to lease at fair-market rents, which is a function of the gap between voucher payment standards and market rents. SOAR — the SSI/SSDI Outreach, Access, and Recovery program — is canonical evidence (Harp et al. 2011, SAMHSA tracking) that benefits-eligible homeless adults can be enrolled at 65 percent approval rates against a 10 to 15 percent baseline through structured advocacy; SOAR is among the highest-leverage workforce investments a CoC can make.^{[f15][workforce]}

4.3 Structural Racism

Black Americans are 13 percent of the U.S. population but 36.6 percent of people experiencing homelessness — a 2.8x overrepresentation that is consistent over time, across regions, and across age groups.^[f9] The disparity is structural, not behavioral. The peer-reviewed literature has identified specific causal pathways: intergenerational wealth gaps rooted in slavery, Jim Crow, redlining, and mass incarceration; housing discrimination documented through HUD audit studies; the criminal justice system as a barrier to housing and employment through background-check policies; and disproportionate cost-burden rates for Black, Hispanic, and Native American renter households relative to White households.^[f9]

The implication for community strategy is that racial equity cannot be a separate initiative — it must be built into the system design at every layer. Coordinated entry assessment tools should be validated across racial subgroups; commonly used tools have not been. Housing programs should remove or mitigate barriers that disproportionately affect people of color: criminal-history background checks, credit-history requirements, documentation requirements that screen out people without continuous identification. Housing placement outcomes should be tracked disaggregated by race and reported at the system level. Workforce diversity matters — representation in clinical and leadership roles has documented effects on client engagement and retention.

The structural advocacy implication is that the racial disparity in homelessness will not be closed by intervention quality alone. It requires direct work on the upstream drivers: criminal justice reform that limits background-check exclusions, housing discrimination enforcement that has been chronically under-resourced, intergenerational wealth-building policy.

4.4 Healthcare and Behavioral Health Gaps

Medicaid expansion is associated with lower homelessness rates at the state level. The mechanism is the access pathway: untreated serious mental illness, untreated substance use disorder, and unmanaged chronic medical conditions are amplifiers of housing risk, and Medicaid access for the population at risk substantially expands treatment access. The state-level association is established across multiple studies; the causal mechanism is indirect but plausible.^[services]

Beyond Medicaid expansion, the healthcare system's role in homelessness has two dimensions. The first is direct: hospitals discharging people experiencing homelessness into the homelessness system without a placement plan produce predictable returns to homelessness and high-cost re-utilization. Critical Time Intervention is the evidence-based response to this — RCT evidence from Susser, Herman, and colleagues established CTI as canonical for reducing return to homelessness at 18 months following institutional discharge.^[services] The second is financial: Medicaid is the fastest-growing and most structurally important new funding lever in homelessness services, with state 1115 waiver authorities (California's CalAIM is the most-developed example) now covering housing navigation, move-in costs, transition support, and time-limited rental assistance in addition to traditional clinical services.^[f15] The administrative capacity to bill Medicaid for these services is uneven and frequently inadequate; building it is among the most consequential operational investments a homelessness agency can make.

4.5 Criminal Justice Exposure

A criminal record creates barriers to housing through screening practices, to employment through hiring policies, and to family stability through the household dissolution that incarceration can produce. Approximately 10 percent of people experiencing homelessness have experienced incarceration in the period surrounding the homelessness episode — an inversion of the conventional framing that pathologizes the homeless individual; the more honest framing is that the criminal justice system is a driver of housing instability for a substantial subset of the homeless population.^[justice]

State and local "fair chance housing" legislation — restricting the use of criminal history in housing applications — is a measurable policy lever. Reentry support that includes housing navigation as a routine component of release planning is another. The federal policy implication is that HUD's screening guidance, last meaningfully updated in 2016, requires modernization to constrain the use of background checks in ways that produce racially disparate outcomes without commensurate safety benefit.

4.6 Safety Net Failures

Homelessness is in important measure the failure point of multiple other safety nets. Foster care youth aging out of the system at 18 or 21 enter homelessness at rates well above the general youth population — Chapin Hall's Voices of Youth Count data documented that approximately 40 percent of youth experiencing homelessness had a foster care history.^[f11][youth] TANF benefits inadequate to cover rent; SSI applications taking years to approve without SOAR-style structured advocacy; veterans discharging without housing plans before the HUD-VASH system reaches them; people leaving jail or prison without reentry housing — each is a safety-net design choice with predictable homelessness consequences.

The implication is that homelessness policy is not separable from social policy more broadly. Extending foster care to age 23, accelerating SSI/SSDI determinations, strengthening reentry housing placement, indexing TANF and SSI to housing costs — these are homelessness policies even when they are not labeled as such.

Part V — The Funding Architecture

The evidence on what works does not automatically translate into investment. The funding architecture is its own structural barrier. Communities that understand the architecture can navigate it; communities that do not produce predictable funding stack failures.

5.1 The Multi-Funder Stack

No single federal program funds the complete set of what a community needs. A working homelessness response combines:

• Capital for PSH and affordable housing development: LIHTC (9 percent and 4 percent with bonds), HOME, state housing trust funds, local bonds, CDFI debt
• Operating subsidy for housing affordability: project-based vouchers, CoC rental assistance, HUD-VASH for veterans
• Services for the supportive layer in PSH and intensive case management: CoC grant services component, Medicaid, SAMHSA PATH, state mental health authorities
• Prevention and emergency rental assistance: ESG, CDBG, TANF, state general fund, philanthropy
• System infrastructure (HMIS, coordinated entry, by-name lists): CoC infrastructure dollars, local general fund
• Workforce and organizational capacity: philanthropy, general operating support, Medicaid billing revenue

The most common funding-stack failure is PSH developed with insufficient services — capital was secured, the building was built, but the services budget never came together. This is the operational manifestation of the cross-agency funding problem and the primary mechanism by which evidence-based PSH becomes underperforming subsidized housing.^[f17]

5.2 The Cross-Budget Problem

The Culhane evidence on cost-offset is canonical: approximately $12,146 per PSH placement per year in service-cost reductions, against $17,277 in annual housing costs.^[f4] But the savings are not distributed proportionally to the agencies that fund the housing. They accrue to OMH (state mental health), Medicaid (state-federal), VA, state and city corrections, and shelter — different agencies, different legislative appropriations, different levels of government than the housing budget.^[f17] No automatic mechanism routes the savings back into housing. This is the structural paradox: 20-plus years of stronger evidence has not unlocked proportional funding because the evidence problem is not a knowledge problem but a budget-structure problem.

Three solutions have emerged or are emerging.

The first is cross-system cost accounting: a community that can document the total public service cost of its highest-utilizer chronically homeless population — across shelter, ED, hospital, jail, mental health — produces a cross-agency ROI conversation that no single-budget figure can produce. The work is operational (governance, data-sharing agreements, analytical capacity) but the return on the investment is the credibility of the cross-agency ask.

The second is single-payer routing through Medicaid. State 1115 waivers (California's CalAIM is the most-developed example, with peer-reviewed evaluation pending) allow Medicaid to fund housing-related services — navigation, transition costs, time-limited rental assistance in some states — directly. When Medicaid funds the services and Medicaid receives the savings, the cross-agency problem is partially resolved. The federal government cannot use Medicaid to fund rent itself, but the surrounding services and transition costs can be Medicaid-financed.

The third is healthcare-system partnership. Hospitals and managed care organizations in capitated arrangements have a direct financial incentive to fund housing for high-utilizer patients whose hospitalizations they bear the cost of. The hospital-CoC partnerships that route Medicaid or hospital savings into housing are growing — and represent one of the highest-leverage funding strategies available to a community whose hospital systems are organizationally sophisticated.

5.3 The Medicaid Opportunity

Medicaid is the structurally most important new lever in homelessness funding. The federal authority exists — Section 1115 waivers and Section 1915(c) home and community-based services authority — and CMS guidance issued in 2022 and 2023 expanded the clarity on health-related social needs (HRSN) services that Medicaid can cover.^[f15] States that have moved most aggressively — California (CalAIM), Oregon, Washington, Washington DC — are building infrastructure that states without aggressive Medicaid action are not.

The barrier is administrative capacity. Medicaid billing for housing-adjacent services requires documentation, billing infrastructure, compliance infrastructure, and clinical credentialing that most homelessness agencies do not have. The states moving aggressively are also having to build the provider capacity to participate in the new authority. The administrative complexity is real and frequently underestimated in policy advocacy that focuses on the regulatory authority without naming the operational ramp.

5.4 Philanthropy and Its Distinctive Role

Foundations and philanthropy occupy a distinctive position in the funding architecture: they fund what federal programs cannot, they take risks federal programs cannot, and they can move faster than federal appropriations. The highest-leverage philanthropic investments are: workforce capacity (general operating support to retain staff and pay competitive wages), organizational infrastructure (HMIS data quality, cross-system data-sharing capacity, evaluation infrastructure), prevention at scale (statistical targeting plus financial assistance, where federal funding is patchy), and policy advocacy (which federal grants restrict but philanthropy does not).

The under-recognized philanthropic role is in evidence infrastructure. Federal funding for homelessness research has not kept pace with the policy attention to the issue. HUD's Office of Policy Development and Research and the Office of the Assistant Secretary for Planning and Evaluation at HHS produce essential evidence but operate under resource constraints. Philanthropic investment in independent evaluation — particularly RCT and quasi-experimental work on diversion, prevention targeting, and PSH at scale — produces the evidence that anchors the next decade of policy. The Notre Dame / J-PAL diversion RCT, expected results 2026–2027, is exactly this kind of investment. More are needed.

5.5 What Funder Discipline Looks Like

The hardest finding in the funder landscape is that not all current philanthropic homelessness investment is well-allocated relative to the evidence. Programs without rigorous outcome data continue to be funded. Interventions whose effect on homelessness is theoretically appealing but evidentially weak — particularly some workforce, employment, and wraparound programs marketed as preventing homelessness — receive substantial allocation. Programs with strong evidence — targeted prevention, PSH with appropriate services, voucher access expansion — are under-resourced relative to the magnitude of the evidence supporting them.

The funder discipline this document is asking for is simple in concept and difficult in practice: allocate the next dollar of philanthropic investment to interventions whose effect on homelessness has been demonstrated at the level the evidence supports. Require evidence standards in grant criteria. Fund evaluation when funding programs. Stop funding interventions whose evidence is sentimental rather than empirical. This is not a critique of any single funder; it is a critique of the aggregate allocation pattern. The aggregate allocation can change as funder norms change, and the leverage on aggregate allocation is the explicit standards adopted by major program officers and trustees.

Part VI — The Policy Agenda

Operational excellence is necessary. Structural change is necessary. Both must happen simultaneously, and the policy agenda is how the structural side gets done. This section names what communities, states, and federal policymakers should be pursuing.

6.1 Federal Policy

The highest-leverage federal investments, in priority order:

Expand the Housing Choice Voucher program. The voucher program is rationed by appropriation. The rationing is the policy choice that produces 39 percent year-over-year increases in family homelessness. Expansion at the scale of need — universal voucher provision for all income-eligible households, phased over a decade — is the single highest-leverage federal investment available. Interim expansion at the scale of the homeless population is achievable within current budget windows; the political question is whether it is prioritized.

Maintain and modernize Housing First as federal policy. HUD's CoC competitive grant scoring should continue to reward Housing First fidelity. The 2024–2026 federal policy environment is creating pressure to reframe federal funding around treatment-first or sobriety-contingent models. The evidence does not support that pivot. The CoC NOFO and the broader federal posture should hold Housing First as a non-negotiable standard.

Fund deep affordability production at scale. LIHTC expansion alone does not address the deep affordability gap. A national rental subsidy or deep-subsidy supplemental program is needed, layered on LIHTC to reach 30 percent AMI. The National Housing Trust Fund exists in statute; the appropriations have not matched the design.

Stand up Medicaid authorities for housing services across all fifty states. CMS has expanded the regulatory pathway through HRSN guidance. State adoption is uneven. A federal technical assistance program to support all fifty states in building Medicaid HRSN infrastructure — modeled on the diffusion of Medicaid managed care over the past two decades — accelerates the cross-agency funding solution.

Modernize HUD screening guidance. The 2016 HUD guidance on criminal background screening is the most recent meaningful update. Stronger guidance is needed to constrain background-check practices that produce racially disparate outcomes without commensurate safety benefits, particularly for non-violent offenses years past.

Restore and expand HUD's evidence function. HUD PD&R produces essential evidence under resource constraints. The federal evidence base is necessary infrastructure. Funding it adequately is itself a policy lever, because without it, state and local actors operate without national context.

Release the 2025 AHAR. The single most consequential current data governance failure is the non-release of the 2025 Annual Homelessness Assessment Report. The data exists; it is being withheld. Release of the 2025 AHAR — and a commitment to annual release going forward — is a basic prerequisite for evidence-based federal policy. The current absence is corrosive to every downstream decision.

6.2 State Policy

Expand emergency rental assistance with statistical targeting. States that have invested in steady-state ERA capacity outperform those that have not. California's HHAP demonstrates what scaled state investment can do, and the evidence supports its expansion to other states. State investment should include statistical targeting infrastructure, not just rental assistance dollars.

Adopt Section 1115 Medicaid waivers for housing services. California, Oregon, Washington, Arizona, and the District of Columbia have moved aggressively. Other states have not. State-by-state adoption is the path to nationwide HRSN capacity.

Reform LIHTC Qualified Allocation Plans. States control LIHTC scoring through their QAPs. QAPs that reward PSH set-asides, deep affordability, and Housing First fidelity steer the largest affordable housing production tool in the country. CoC engagement with state QAP comment processes is one of the highest-leverage advocacy investments available to a homelessness coalition.

Fund right-to-counsel in eviction proceedings. Right-to-counsel in eviction proceedings is among the highest-ROI prevention investments available through legislation. New York City data showed a 41 percent reduction in evictions and 86 percent of tenants remaining housed; San Francisco's program reported 92 percent of tenants avoiding homelessness.^[f13][pl3] Federal authority does not exist; state and local action is the available lever.

Pass fair chance housing legislation. State-level fair chance housing legislation restricts the use of criminal history in housing applications. Multiple states have passed it. The evidence on disparate-impact effects of background checks is sufficient to support broader adoption.

Extend extended foster care to age 23. Foster-care-to-homelessness is one of the most predictable pipelines in the system. Extending extended foster care to age 23 — with housing support, education, and benefit access — narrows the funnel. State action exists; federal Title IV-E support is available.

6.3 CoC and Local Policy

Adopt a community by-name list and operate it in real time. This is the single most important local commitment a CoC can make. It is operational, not legislative, and within the authority of CoC leadership without state or federal action.

Run coordinated entry to evidence-based standards. Acuity-driven prioritization, real-time list management, racial equity audit, diversion at intake, integration with HMIS. Federal CoC standards permit this; the failure is implementation, not authority.

Build PSH pipeline to identified need. The chronically homeless population in each community is identifiable through the by-name list. The PSH pipeline can be sized to that need with five-year construction and acquisition timelines. The work is housing development capacity, financing strategy, and local political will.

Expand voucher access strategy. Active PHA partnership, project-based voucher commitments for new development, supplemental state or local rental assistance. CoCs that treat the voucher pool as fixed are ceding the leverage. CoCs that build voucher access strategy as a core function unlock the binding constraint for family homelessness.

Track racial equity at every layer. Coordinated entry, prioritization, time-to-housing, return-to-homelessness — all disaggregated by race, reviewed at governance, with disparities surfaced and addressed. This is governance practice, not a separate initiative.

Build cross-system cost accounting capacity. The credibility of cross-agency funding asks depends on the data. Partner with hospital systems, state Medicaid, county jail, and state mental health to build the data infrastructure. The reward is the funding conversation that follows.

Invest in workforce. Pay above sector median. Staff intensive programs at evidence-based caseload ratios. Build supervision, clinical consultation, debriefing, mental health benefits, career pathways. Recruit from the communities served. The workforce is the service quality multiplier; treating it as a fixed cost is a strategic error.

Part VII — Functional Zero and What It Demands of a Community

7.1 The Definition

Functional zero is a community state in which homelessness is rare, brief, and non-recurring. The operational definition the field has converged on: the number of people experiencing homelessness in any given month is at or below the system's monthly housing placement capacity. If 15 people become homeless in a month and the system houses 15 people in that month, the count stays at zero or decreases. People who become homeless are identified quickly, housed quickly, and do not return.^{[functional_zero]}

The definition matters because it shifts the goal from "housing everyone who is currently homeless" to "building a system capacity that exceeds inflow." It is a flow target, not a stock target. A community can house thousands of people in a year and not reach functional zero if inflow exceeds the placement capacity. A smaller community can reach functional zero with modest placement capacity if inflow is correspondingly small. The definition is local, scaled to the community's conditions.

7.2 The Veterans Proof of Concept

The veterans population is the field's most compelling proof that functional zero is achievable in the United States. National veteran homelessness has declined by approximately 55 percent from 2009 to 2024, falling to about 32,800 on the January 2024 night.^[f10] Twelve or more communities have documented functional zero for veterans, including Douglas County, Colorado (functional zero achieved September 2024, with 64 percent of veterans housed within 98 days of identification) and Hennepin County, Minnesota (167 homeless veterans in August 2023, down to 69 by September 2024).

The transferable elements: dedicated subsidy pool (HUD-VASH vouchers, not competing from the general HCV pool); integrated services (VA provides healthcare, behavioral health, and case management through a single system); by-name data (every veteran experiencing homelessness tracked by name in real time); functional zero as the explicit target, not "reduction"; named public accountability (specific leaders responsible for the outcome in specific communities).

The non-transferable element: veterans have a cross-partisan political constituency that other populations lack. Replicating the infrastructure for non-veteran populations requires building political will that does not arise automatically. This is a strategic problem, not a technical one. The technical model is replicable.

7.3 What Functional Zero Requires

Communities that have achieved or approached functional zero share a consistent set of conditions:

Political will with named accountability. An elected official or senior administrator whose professional reputation is attached to the outcome. Diffuse accountability produces diffuse results.

A single leader of the homeless response system. Not a committee, not a loose coalition, but a person who can make decisions and be held responsible. Most communities have committees; some have leaders.

Real-time data. By-name list updated in real time. Leaders can say on any given day exactly how many people are experiencing homelessness in the community, by name, and where each is in the pipeline.

Housing supply meeting identified need. Enough PSH beds, vouchers, and affordable units to house everyone identified as needing housing. This is the binding constraint in most communities and the prerequisite that all the operational excellence in the world cannot substitute for.

A services system with fidelity. Case management, behavioral health, and peer support delivered at the intensity the evidence supports, with staff retained and supported.

A sustained funding commitment. Not a one-time surge but ongoing investment. Functional zero requires maintenance even after it is achieved. Communities that stop investing in data and operations backslide.

7.4 The Timeline

The reasonable timeframe for a well-positioned community to reach functional zero for chronic homelessness is three to five years. For families with adequate voucher access, similar. For overall functional zero across all populations, longer — the supply and income constraints take a decade-plus to resolve, and the political constituency for full system reform takes years to build.

The dangerous claim — "homelessness can be ended in eighteen months" — sets up failure. The honest claim — "homelessness for a defined population can be reduced to rare, brief, and non-recurring within five years if the operational and supply conditions are met" — is defensible and achievable. The framing matters for funder expectations, political durability, and community morale.

Part VIII — The Asks

This is the section the rest of the document has been arguing for. Each ask is directed at a specific audience.

8.1 To Federal Policymakers

Expand the Housing Choice Voucher program toward universal provision. The rationing is the policy choice that produces the family homelessness increase.

Hold Housing First as the federal evidence standard. The CoC NOFO scoring, HUD technical guidance, and federal posture should not be reframed around treatment-first or sobriety-contingent models. The evidence does not support the reframing.

Release the 2025 AHAR and commit to annual release. No evidence-based federal policy is possible without national outcome data.

Fund deep affordability at the scale of the supply gap. National Housing Trust Fund expansion and capitalization, layered with operating subsidy to reach 30 percent AMI.

Stand up Medicaid HRSN authorities in all fifty states. CMS technical assistance program for state implementation.

Fund HUD PD&R and HHS ASPE adequately. The federal evidence base is infrastructure. It is also under-resourced.

8.2 To State Policymakers and Housing Finance Agencies

Expand state-funded emergency rental assistance with statistical targeting. California's HHAP and the Evans/Phillips evidence on targeted prevention together make this the highest-leverage state investment available.

Adopt Section 1115 Medicaid waivers for housing services. California, Oregon, Washington, Arizona, and DC are the model. Replicate.

Reform QAPs to reward PSH, deep affordability, and Housing First fidelity. The largest affordable housing production tool in the country is in state hands.

Fund right-to-counsel in eviction proceedings. Among the highest-ROI prevention investments available through legislation.

Pass fair chance housing legislation. Restrict criminal-history use in housing applications.

Extend extended foster care to age 23. Close the foster-care-to-homelessness pipeline.

8.3 To CoC Leadership and Local Government

Adopt a real-time by-name list as the operational backbone of the system. If you do nothing else from this document, do this.

Run coordinated entry to evidence-based standards. Acuity-driven prioritization, racial equity audit, diversion at intake.

Build PSH pipeline to identified need on a five-year timeline. Use the by-name list to size the chronically homeless population; build to it.

Expand voucher access strategy. Active PHA partnership, project-based voucher commitments, supplemental rental assistance.

Track racial equity at every layer. Disaggregate by race; review at governance; act on the gaps.

Build cross-system cost accounting capacity. The credibility of the funding conversation depends on the data.

Invest in your workforce above the sector median. The workforce is the service quality multiplier; treat it as such.

8.4 To Foundation Program Officers and Trustees

Allocate the next dollar to interventions whose effect is established at the level the evidence supports. This means more prevention with targeting, more PSH with services, more voucher access expansion, more workforce capacity, more evaluation. It means less funding of interventions whose evidence is sentimental rather than empirical.

Adopt evidence standards in grant criteria. Specify in RFPs what evidence base programs must reference. Require outcome reporting commensurate with funding scale.

Fund evaluation when you fund programs. The Notre Dame / J-PAL diversion RCT is exactly the kind of investment philanthropy is best positioned to make. There should be more.

Fund workforce capacity, not just program delivery. General operating support that retains experienced staff is among the most leveraged philanthropic investments available.

Fund policy advocacy. What federal grants restrict, philanthropy can support. The cross-agency funding solution, voucher expansion, and Medicaid HRSN diffusion all require advocacy infrastructure that program dollars do not fund.

8.5 To Healthcare and Managed Care Leadership

Build hospital-CoC partnerships that route Medicaid savings into housing. Your highest-utilizer patients are also the population PSH most reliably stabilizes. The financial alignment exists; the operational partnership has to be built.

Adopt Medicaid HRSN authorities at the MCO level where state policy permits. Housing navigation, transition support, and ancillary services for the population whose hospitalizations you are bearing the cost of.

Invest in cross-system data sharing. The data infrastructure that enables cross-agency cost accounting is in your network — hospitals have the data the CoC needs and the technical capacity to share it. Initiating the partnership is the leverage point.

Part IX — What This Document Is Not Saying

A document that argues a position should be explicit about what it is not arguing.

This document is not arguing that homelessness is solely a structural problem and individuals have no agency. The evidence on Housing First does not say behavioral health is irrelevant; it says housing precedes treatment, and treatment is essential for the highest-acuity populations. Individuals make choices within structures; the structures shape the choices available. Both are real.

This document is not arguing that all current homelessness spending is well-allocated. Some is. Some is not. Funder discipline — allocating the next dollar to the evidence-supported intervention — is the lever for changing the aggregate allocation. This is a critique of patterns, not of individuals.

This document is not arguing that the federal political environment is irrelevant to outcomes. It is. Federal policy choices over the next four years will materially shape what communities can achieve. Communities should not wait for federal policy to align, and federal policy should not be the determinant of local possibility — but the federal layer is not optional.

This document is not arguing that Housing First is sufficient on its own. Housing First is the framework; PSH-with-services is the program; the operational layer is the implementation; the structural conditions are the inflow. All four matter. A community that does only one will produce predictable failure.

This document is not arguing that the United States can solve homelessness in three years. The supply and income constraints will take a decade-plus to resolve. The operational layer can move in three to five years for defined populations. Functional zero for a community across all populations is a longer horizon. Naming this honestly is the precondition for sustaining the political will to do the work.

Appendix A — Definitions

By-name list

A real-time, individual-level registry of every person experiencing homelessness in a defined community, with current status, time in system, and assigned navigator.

Chronic homelessness

Defined by HUD as homelessness lasting at least 12 continuous months (or 4 episodes in 3 years totaling 12 months) combined with a disabling condition.

Continuum of Care (CoC)

The geographically defined regional planning body designated by HUD to coordinate homelessness services and receive federal CoC funding.

Coordinated Entry (CE)

The federally required system through which a CoC routes people from intake to appropriate intervention.

Functional zero

A community state in which homelessness is rare, brief, and non-recurring, operationalized as: monthly homelessness inflow at or below monthly housing placement capacity for a defined population.

HMIS

Homeless Management Information System — the federally mandated data system for tracking homelessness service utilization at the CoC level.

Housing First

A framework prioritizing immediate access to permanent housing without preconditions of sobriety, treatment compliance, or "housing readiness."

PSH (Permanent Supportive Housing)

Permanent housing with affordability subsidy combined with onsite or wraparound supportive services for chronically homeless or disabled populations.

RRH (Rapid Rehousing)

Time-limited rental assistance and supportive services to move households quickly out of homelessness, typically 3 to 24 months of subsidy.

System Performance Measures (SPMs)

HUD-defined community-level metrics on homelessness system performance, including length of time homeless, returns to homelessness, exits to permanent housing, and others.

Evidence Index

The following index maps every claim in this document to its anchor finding in the underlying knowledge base. The knowledge base is the authoritative source. Where this document and the knowledge base diverge, the knowledge base is correct.