Summary
What do they do? SF Trust Youth Initiative combined two years of unconditional payments with optional supportive services for young adults experiencing homelessness. Larkin announced completion in December 2025 and named Accelerating Independent Missions (AIM) as the next initiative. We do not assume AIM has the same dose, costs or effects. More
Why this approach interests us
- Flexible cash plus optional support can pay deposits, bridge income gaps and let young people choose housing arrangements.
Our main reservations
- The available results do not identify additional housed person-years or a causal health-utility gain. Housing is not automatically attributable to cash.
What do you get for your dollar?
Our central scenario is $40.4M per better life (10 QALYs). It models a new program at the completed Trust Youth Initiative’s cash dose—not a measured Larkin effect or an available donation offer. The housing and health inputs are explicit best guesses. Inspect the full model →
- our best guess
- $40.4M — per 10 incremental QALYs; conditional health-only scenario
- positive scenarios
- $6.07M–$2.42B — Judgmental stress tests, not confidence bounds
- cash plus support
- $50,500 — Published $40,500 dose + assumed $10,000 support
- funding room
- Unverified — Completed pilot; successor economics unknown
1. What do they do?
SF Trust Youth Initiative combined two years of unconditional payments with optional supportive services for young adults experiencing homelessness. Larkin announced completion in December 2025 and named Accelerating Independent Missions (AIM) as the next initiative. We do not assume AIM has the same dose, costs or effects.
Finance a genuinely new cohort
Separate new cash and staff capacity from replacement of existing funding.
Offer flexible cash and support
The historic dose was $1,500 monthly for 24 months plus $4,500.
Compare housing trajectories
Count additional stable-housing days against ordinary services, then apply a transparent health-only bridge.
Scope of this review. TYI, AIM, federal RHY prevention and residential housing are separate until a program/funding crosswalk proves otherwise. Do not divide Larkin’s whole budget by every young person served.
2. Monitoring and information sharing
85 young adults; 45 TYI and 40 usual-care enrollment groups
Primary SF study design. Three initially randomized invitees were reassigned after nonresponse.
Our assessment. Enrollment-group comparisons are not clean intention-to-treat estimates. Severe untreated mental-health/substance-use conditions were excluded.
SF pilot participants
Conference abstract and organization summary. The abstract uses paired baseline-to-month-18 tests; the organization reports 86% permanently housed at 24 months.
Our assessment. Neither result supplies the cumulative causal housing-days used by this model. The 86% denominator/comparator is not supplied in the announcement.
NYC young adults; different dose and housing system
External cash-plus analogue. Published enrollment-group trajectories suggest benefit but include reassignment and differential response.
Our assessment. A difference in any homelessness during one month is not the same as additional housed years. Not used as a numeric effect input.
3. Qualitative assessment
Flexible cash plus optional support can pay deposits, bridge income gaps and let young people choose housing arrangements.
Key reservations
- Housing-to-health utility is subjective and correlated with other housing models.
- Causal housing duration and missing follow-ups remain unresolved.
- Whole-organization financial pressure is not program-specific funding room.
- AIM and other prevention programs could have different economics.
Benefits not included in our estimate
- Mortality and health beyond 30 months
- Income, autonomy, education and justice benefits outside the health-utility bridge
- Government savings and transfer recipients' cash value as separate QALYs
- Other Larkin housing, employment and engagement programs
4. What do you get for your dollar?
Our best estimate: $40.4M per 10 QALYs.
We assume six additional months housed over 30 months, a 0.05 health-utility improvement during that time, and 50% donor-funded additional capacity. These are deliberately exposed analyst judgments, not trial results. The $10,000 support assumption covers 96 loaded staff hours at $100 plus $400 administration.
A better life is our comparison unit of 10 additional quality-adjusted life years (QALYs), potentially spread across people. These are uncertain estimates, not measured returns or verified donation offers.
How we calculate the estimate
DOLLARS PER BETTER LIFE: 10 × $50,500 ÷ (0.5 years × 0.05 × 50%)
$40,400,000 per 10 QALYs
Model inputs and assumptions
- Cash per participant
- 40500 (range: 40500 / 40500 / 40500). Published SF TYI entitlement: $1,500 monthly for 24 months plus $4,500 = $40,500. Historic nominal dose retained without inflation adjustment; not an AIM quote. Published cash dose; otherwise explicit judgment.
- Support and administration
- 10000 (range: 5000 / 10000 / 20000). Central assumed $10,000: 96 staff hours over 24 months at an assumed $100 loaded hourly cost, plus $400 administration. Includes unsuccessful engagement; not observed Larkin spending. Positive stress cases $5,000/$20,000. Published cash dose; otherwise explicit judgment.
- Extra stable-housing person-years
- 0.5 (range: 1 / 0.5 / 0.1). Judgment: half a year newly housed over 30 months versus ordinary services, equivalent to 20% of the observed window. This gives partial rather than near-total credit because ordinary services also improve housing; it is a provisional decision assumption, not a calibrated empirical estimate. Stress cases 0.1–1 year. Not derived from the reported 86% housing rate, before–after nights, or NYC enrollment-group difference; clean cumulative causal housing-day data are unavailable. Published cash dose; otherwise explicit judgment.
- Health utility per additional housed year
- 0.05 (range: 0.1 / 0.05 / 0.01). Explicit subjective 0.05 (0.01–0.10): a modest health-related quality improvement of five percentage points during extra housed time, not credit for all housing benefits, not measured by TYI and not a validated homelessness-to-QALY conversion. It credits only health-related quality of life; shares uncertainty with other housing models. Published cash dose; otherwise explicit judgment.
- Additional donor-financed capacity
- 0.5 (range: 0.75 / 0.5 / 0.25). 50% judgment (25–75%) for replacement of already financed cash/support. Housing effect above is already relative to services as usual; do not apply a second causal housing discount. Published cash dose; otherwise explicit judgment.
Illustrative $100,000: 1.98 packages and 0.0248 additional QALYs
- Favorable positive: $6.07M per 10 QALYs. 0.165 QALYs per $100,000; not available funding room
- Central analyst estimate: $40.4M per 10 QALYs. 0.0248 QALYs per $100,000; not available funding room
- Unfavorable positive: $2.42B per 10 QALYs. 0.000413 QALYs per $100,000; not available funding room
Uncertainty. Zero extra housing, zero utility gain, or full funding substitution yields no finite positive-benefit price. Harm remains possible and is not assigned a positive price. Positive scenarios are judgmental stress tests, not confidence bounds. The 30-month horizon excludes later effects.
Could this beat $100,000 per 10 QALYs?
The cash alone would require more than 4.05 additional QALYs per participant. In a deliberately extreme 30-month health-only ceiling of one extra QALY per year, that is impossible: even the cash-only price is $162,000 per 10 QALYs. This ceiling assumes utility gains between zero and one and no mortality or later effects; it is not a universal limit on cash benefits.
CASH-ONLY THRESHOLD: 10 × $40,500 ÷ 2.5 QALYs
$162,000 even at this extreme short-horizon ceiling
QALY conversion assumptions
Longer durable benefits, mortality effects, lower-cost interventions or benefits outside health could change the conclusion. They need separate evidence, not silently extended follow-up.
5. Funding and previous grants
No priced additional replication or verified funding gap. AIM may differ from the completed pilot.
FY2025 audited government-contract revenue was $22.58M and total expenses $36.78M. These whole-organization accruals neither price TYI nor establish an unfunded tranche. We do not net presumed public savings out of donor cost.
This review does not establish a verified marginal funding offer or a complete history of grants.
We have not verified a suitable donation route for this reviewed activity. Confirm the legal recipient and intended allocation before donating.
6. Sources
- SF Pathways Study and Trust Youth Initiative overview. Chapin Hall. Primary design and enrollment report; 85 participants, three reassignments; not final intention-to-treat outcomes. Published: 2025-03; retrieved: 2026-09-07.
- Implementing Cash Transfers in the Context of Local Homelessness Response Systems. SSWR / Sara Semborski. Primary conference abstract: cash dose and paired before–after housing analysis, not a clean between-arm effect. Published: 2026-01 conference; retrieved: 2026-09-07.
- Celebrating the Trust Youth Initiative. Larkin Street Youth Services. Organization-reported 86% housing outcome and AIM successor announcement; denominator and counterfactual not supplied. Published: 2025-12-15; retrieved: 2026-09-07.
- NYC Pathways Study Findings. Chapin Hall. External cash-plus analogue; reassignment and differential response limit causal inference; not an effect input. Published: 2025-07; retrieved: 2026-09-07.
- FY2025 audited financial statements, printed page 5. Larkin Street / independent auditors. Organization-wide financial context; table visually checked; no marginal cash-program cost. Published: 2025-12-19 audit; website path May 2026; retrieved: 2026-09-07.
Annual expenses: years and sources
Average annual expenses (three consecutive fiscal years): $36,262,730. Organization size is separate from the modeled cost-effectiveness of a donation.
Larkin Street Youth Services
Whole legal entity Form 990 Part IX total functional expenses; latest original reconstructed filings checked, fiscal-end year from printed reporting dates. Older fallback rows from IRS extracted API.
Latest original filings checked; ProPublica extracted API lags these original returns.