Summary
What do they do? EIN94-3019992, formerly Chronicle Season of Sharing Fund. Nine Bay counties use coordinating agencies and fiscal agents to pay landlords, utilities and other vendors. Gifts also support food banks. More
Why this approach interests us
- Timely assistance can avert some homelessness; a Santa Clara randomized trial supports the causal housing pathway.
Our main reservations
- Prevented homelessness is not a QALY. Finite health benefits and funding response are judgments;96.3% of expected health comes from the favorable world.
What do you get for your dollar?
The modeled housing-health component of a $10,000 whole gift yields 0.01326 expected Bay QALYs, or $7,541,069 per10. This is not measured health impact or a complete estimate of food-related benefits. Inspect the model, assumptions and scenarios.
- weighted bay cost per10 modeled qalys
- $7,541,069 — Whole cash expense; housing-health component only. Foodhealth unquantified.
- whole fy2025 expense proxy
- $14,799,462 — Audited resource expense$15,130,343 includes$330,881 donated advertising.
- favorable-tail dependence
- 96.3% — Subjective scenario mass below$1M/10:0%; not a confidence interval.
1. What do they do?
EIN94-3019992, formerly Chronicle Season of Sharing Fund. Nine Bay counties use coordinating agencies and fiscal agents to pay landlords, utilities and other vendors. Gifts also support food banks.
Charge whole operations
Keep sponsor-funded administration, fundraising, agent/county fees and food-bank allocations in whole expense; the100%-to-help claim is not zero overhead.
Deliver assistance beyond funded baseline
Model cash additionality for reserves, other donors/public sources and partner capacity. Convert the housing/family pool into grant equivalents, not observed unique households.
Convert only incremental housing to finite health
Absolute risk difference × transfer × effective affected persons × utility × avoided duration; independently subtract harms.
Scope of this review. No food-bank meal-to-QALY credit, no duplicated food recipient benefits. Food health is omitted, not asserted zero. Full societal costs of partnercasework/volunteering remain incomplete. No lifetime, mortality, income or fiscal-savings credit.
2. Monitoring and information sharing
At-risk applicants, not all nine-county grantrecipients
Santa Clara randomized assistance offer,578observations. Six-month adjusted ITT reduction3.8percentagepoints from4.1%control homelessness; rawtreatment0.9%.
Our assessment. Administrative homelessness outcome, not measuredQALY. Grantrecipient effect is not automatically ITT; usingofferanchor with transfer is explicit conservative convention.
Pandemicprogram applicants under different housingpolicies
Emergency rental-assistance lottery evidence, fourcities. Rentpayment/eviction-concern benefits with suggestive self-reportedhealth effects.
Our assessment. Does not establish the model’sutility orduration. No symptom-score-to-QALY mapping invented.
Whole organizationFY2025
Original auditedaccounts and990. Supportcosts real despite sponsors covering them; foodandhousing grants distinct.
Our assessment. No program-only or netfiscal-savings headline. Partner service costs incomplete.
578 pre-March2020 applicants; not the full pandemic sample
Santa Clara trial Table2 and TableA6; independent source audit. The .038 estimate is agency-month-strata-adjusted ITT. Effects were absent during the pandemic and stronger without children. Table2 reports 2.5 fewer recorded homeless-days at6months and7.5 at12months; covariate-adjusted TableA6 estimates are1.9 and5.7days.
Our assessment. The untransferred .038 risk reduction × .5year assumption implies6.94days peroffer, exceeding the six-month2.5day estimate. Broader unrecorded health duration could differ, but is not established. These empirical duration anchors do not validate the assumed utility or household multiplier; coefficients unchanged.
3. Qualitative assessment
Timely assistance can avert some homelessness; a Santa Clara randomized trial supports the causal housing pathway.
Key reservations
- Health conversion is much weaker than housing prevention evidence.
- Grant size, unique-recipient dedup and housing-specific allocation are priors; historical rent-month and family counts do not resolve them.
- Foodhealth omitted; this quantifiedhousingcomponent is not a complete whole-org health valuation.
- Favorable high-risk transfer and householdhealth multiplier dominate expectation.
- No proven marginal funding tranche or guarantee of100000dollar scaling.
- Same public/partner outcomes must not also be credited in full to recipient agencies or foodbanks.
Benefits not included in our estimate
- Food-relatedhealth remains unquantified; no mealcount conversion.
- No long-run mortality, childdevelopment, income, legal or crime benefit added.
- No avoided shelter/health-systemcosts subtracted from donatedcash.
4. What do you get for your dollar?
Whole-gift housing-health model
Nearly5000families is an approximate activity statement.4482is months of rent, not unique families. The model uses housing/family dollars divided by an assumed grant size with dedup and housing relevance; no unsupported exact annual patient denominator.
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
Net Bay housing health: gift/whole expense × ((family/housing pool − provider grants)/assumed grant × dedup) × cash additionality × [housing relevance × .038 × transfer × persons × utility × years − harmQ] × Bay share
0.013260718850430987 expected Bay QALYs per$10000
Model inputs and assumptions
- Whole expense
- $14,799,462 (range: $15,130,343 with disclosed in-kind). IRS/audit reconciliation; sponsor operatingcosts retained. High forFY2025.
- Grant-equivalent denominator
- 4,248 annual equivalents (range: 2,875–5,478). $11,179,746 family/housing cash; $2,000–$3,500 assumed grant and90–98%dedup. Not observed distinct households. Low.
- Housing-relevant share
- 75% (range: 50–90%). Pool includes critical needs; not every grant prevents homelessness. Low.
- Gift-created assistance
- 70% (range: 0–90%). Reserves and partner/publicfunding are alternatives; no dated incremental tranche. Very low.
- Six-month homelessness risk difference
- .038 × .75 (range: Transfer.4–1.5). Adjusted ITT, not relative reduction or mortality. Favorable transfer>1 is an explicit high-risk extrapolation. Strong external trial; uncertain transfer.
- Persons × utility × incremental years
- 1.5 × .10 × .50 (range: 1–2.5 × .05–.20 × .25–1year). All analystjudgments; effective health-person equivalents are not measuredhouseholdsize. Duration ends with alternativehousing/convergence. Recorded homeless-days provide a narrower empirical anchor:2.5days avoided peroffer at6months versus6.94days implied by .038×.5years before transfer. Health persistence beyond recorded homelessness is unverified. Very low.
- Independent harmQ per delivered grant
- .0002 (range: 0–.003). Potential delayed alternativecare, misallocation or instability; not measured adverseevents. Very low.
- Bay health share
- 99% (range: 97–100%). Nine-county program; exact residence at healthbenefit time unavailable. High scope, uncertain migration.
$10,000 unrestricted gift; no linear large-scale assurance
- funding_null: Weight0.15; Not defined per10. BayQ0; full inputmatrix in model and results. Signed/null cases retained.
- health_null: Weight0.15; Not defined per10. BayQ0; full inputmatrix in model and results. Signed/null cases retained.
- harm: Weight0.05; Not defined per10. BayQ-0.00596793514177745; full inputmatrix in model and results. Signed/null cases retained.
- cautious: Weight0.2; Not defined per10. BayQ-0.0002289330456446901; full inputmatrix in model and results. Signed/null cases retained.
- central: Weight0.3; $35,826,204 per10. BayQ0.0027912529986021616; full inputmatrix in model and results. Signed/null cases retained.
- favorable: Weight0.15; $1,174,856 per10. BayQ0.08511684211378766; full inputmatrix in model and results. Signed/null cases retained.
- audited_resources: Include disclosed donated advertising. $7,709,670 per10; BayQ0.012970724108477714. Base coefficients unchanged.
- program_only_comparison: Sponsor-covered support excluded, comparison only. $6,814,626 per10; BayQ0.014674318593025681. Base coefficients unchanged.
- full_cash: Cash additionality100%. $6,492,826 per10; BayQ0.015401614167985429. Base coefficients unchanged.
- no_favorable: Remove favorable world and renormalize. $172,346,486 per10; BayQ0.000580226509838633. Base coefficients unchanged.
- one_person: One health-beneficiary equivalent per episode. $19,042,871 per10; BayQ0.005251309076774491. Base coefficients unchanged.
- half_health: Half utility and half duration. $34,223,251 per10; BayQ0.002921990047615619. Base coefficients unchanged.
Uncertainty. Subjective weights fixed before execution. No world currently below$1M/10 for quantifiedhousinghealth; this does not establish a zero probability in reality, especially with omittedfoodbenefits.
5. Funding and previous grants
HOLD giving recommendation. Independently audited for exploratory publication; current marginal capacity and the unquantified food-health contribution remain unresolved.
FY2025 cash$5.305M, unrestrictednetassets$5.233M andrestricted$2.664M. Sponsors pay supportingexpenses. These are fundedbaseline, not proof of eitherzero or unlimited room. Some county/agent fees included; other complementary public/partnercosts unknown.
This review does not establish a verified marginal funding offer or a complete history of grants.
Opens the organization’s giving page. A general donation may not fund the specific activity modeled here; confirm allocation with the recipient.
6. Sources
- Identity, sponsor support and nine-county delivery. Season of Sharing Fund. Primary organization. Published: Current; retrieved: 2026-09-11.
- FY2025 audited financial statements. Season of Sharing Fund / Baker Tilly. Primary audited accounts. Published: 2025-12-03; retrieved: 2026-09-11.
- FY2025 Form990. Season of Sharing Fund / IRS. Primary return. Published: FY2025; retrieved: 2026-09-11.
- FY2024–2025 annual report. Season of Sharing Fund. Primary annual report. Published: 2025; retrieved: 2026-09-11.
- Rental and housing assistance. Season of Sharing Fund. Primary organization. Published: Current; retrieved: 2026-09-11.
- Santa Clara homelessness-prevention randomized trial, author manuscript. Phillips and Sullivan. Primary randomized trial. Published: 2023-04; retrieved: 2026-09-11.
- Emergency rental assistance during pandemic, four cities. Collinson et al.. Primary randomized-lottery study, abstract inspected. Published: May2024; revisedOctober2025; retrieved: 2026-09-11.
- Santa Clara program context. Notre Dame researchers. Primary research institution. Published: 2023; retrieved: 2026-09-11.
Annual expenses: years and sources
Average annual expenses (three consecutive fiscal years): $15,205,218. Organization size is separate from the modeled cost-effectiveness of a donation.
Chronicle Season Of Sharing Fund
Form 990/990-EZ reported whole-entity expenses; includes program, administration and fundraising costs, but excludes any costs netted against revenue.
Latest original return and prior-year comparative combined with earlier filing data.