GiveBetter x USA

Food Research & Action Center

Nutrition-program policy, implementation and regranting

Research time: 12 min on GPT-6 Astra Light
  • Research — reviewed programs, finances and impact evidence.

Updated: 2026-09-14

Donate

Summary

What do they do? FRAC works to improve access to federal nutrition assistance through policy research, advocacy, technical assistance and grants. Its portfolio includes WIC, SNAP, school meals and local anti-hunger projects. A provisional SNAP-continuity model suggests about $69 million per 10 USA QALYs, but leaves major WIC and school-meal pathways unpriced and is not a measured return.

Why we’re interested in this organization:

  • Federal nutrition programs reach populations whose health can improve when assistance becomes more reliable.

  • Research, implementation support and regranting provide several plausible routes beyond simply publishing advocacy materials.

  • Original accounts permit recipient-only cost reconciliation, and current policy work identifies a concrete administrative-funding decision.

Our main reservations:

  • The central result depends on subjective avoidable-churn, coalition-leverage and funding-additionality assumptions.

  • Historical lifetime health effects may transfer poorly to short interruptions in today's SNAP; zero health gain remains plausible.

  • Federal WIC regrants dominate recent costs, but the reviewed evidence does not identify their additional enrollment or the next unrestricted dollar's destination.

What do you get for your dollar? $69.5M per better life: ten additional quality-adjusted life years in USA. SNAP continuity health only; other impacts unestimated.

Additional preschool child-years of SNAP receipt; WIC enrollment/retention and school-meal access tracked separately but not converted to health without causal bridges.

1. What do they do?

FRAC is a national nutrition-policy and implementation charity, not a food bank. It supports access to WIC, SNAP, child nutrition and related programs, including DC Hunger Solutions and Maryland Hunger Solutions. The 2024 recipient return reports $13.76 million for early-childhood nutrition, of which $11.99 million was grants, and $1.80 million for child nutrition. These are spending categories, not health results. The separate FRAC Action Council is outside this recipient estimate. Recipient return

WIC CIAO awarded $14.8 million in its second and final subgrant round in January 2025, backed by USDA and implemented with partner organizations. The award announcement does not establish additional enrollment, retention or health. It also does not establish that an ordinary donation will expand the already awarded federal project. WIC CIAO

2. Monitoring and information sharing

The useful outcomes are additional benefit-receipt months, reduced administrative churn, successful enrollment and redemption, with appropriate comparison groups—not the number of policy reports, eligible people or schools offering meals. USDA's 2023 table counts 4.464 million preschool-age SNAP participants nationally, defining preschool as younger than five; the table includes territories. This anchors scale, not FRAC attribution. USDA

For the modeled branch, monitor whether administrative relief changes state staffing and timeliness, whether eligible families actually retain benefits, and whether FRAC changes the outcome relative to other advocates and public agencies. For WIC, request project-level enrollment and retention effects against a credible comparator, unique participants, age/pregnancy mix and cost, separated from general WIC participation trends. The reviewed sources did not supply those decisive marginal results.

3. Qualitative assessment

The clinical bridge is promising but indirect. A large historical rollout study links earlier Food Stamps access to later survival; its preferred treated estimate is about 1.2 additional life-years for exposure from conception to age five. That is model-based life expectancy, not observed completed lifespans, and the design observes early access alongside later access. Modern short gaps may behave differently. Study

A contemporary pre-pregnancy SNAP analysis finds that some favorable conventional estimates do not survive its event-study checks. This discourages simply assigning a fixed infant-health dividend to every benefit month. Contemporary study WIC has a separate evidence base, but associations and program effectiveness alone do not identify FRAC-induced enrollment or QALYs. AHRQ review

The current SNAP administrative-cost decision supplies a concrete prospective route. FRAC argues for relief while USDA's June 2026 proposed rule reflects a 25% federal administrative share from FY 2027. Preventing administrative disruption is not the same as reversing every statutory eligibility cut. State spending could replace federal funding anyway; relief could also redirect public resources from other valuable services. FRAC USDA proposal

4. What do you get for your dollar?

The recipient's 2024 expense is $21,497,588 after restoring $14,674 of event costs netted against revenue; do not use the larger consolidated affiliate total. Two annual budgets, $42,995,176, fund the normalization used below. This includes all recipient programs and overhead, not only SNAP advocacy. 2024 return Comparable gross-expense figures are $14,821,886 in 2023 and $9,720,220 in 2022; the three-year mean is $15,346,565. 2023 return 2022 return

My weak-confidence central judgment uses four million early-child SNAP participants, 5% additional annual benefit continuity for two years conditional on effective relief, a 10% coalition-induced change probability, 5% FRAC attribution and 50% funding additionality. That yields 1,000 expected additional child-years, about $42,995 per child-year. A deliberately attenuated and discounted lifetime bridge gives about 0.0062 QALYs per child-year, then a 99.8% USA resident share: roughly 6.2 USA QALYs, or $69 million per 10. These probabilities and effects are opinions, not measured program performance.

This is a partial-pathway estimate charged full recipient costs, not a proven lower bound or a complete valuation. WIC, school meals, adult nutrition and other policy benefits are centrally net zero solely because they are unpriced; adverse effects can reverse the result. Federal food benefits, state administration and the opportunity cost of taxes or displaced spending are outside the charity-cost numerator. Thus this is not a social cost-effectiveness estimate.

The joint favorable scenario is about $668,000 per 10 USA QALYs; the poor positive scenario is about $584 billion. These are stress tests, not confidence bounds. Zero and negative scenarios are retained. On this narrow branch alone, the central case is not especially compelling; a stronger recommendation would need evidence for additional WIC/school-meal yield or substantially better policy leverage.

Model, assumptions and sensitivity

Two years of full recipient-only gross operating expense, including regrants and overhead, excluding separate active Action Council. Accrual Form 990 basis restores event costs. Public benefits and public administrative/social opportunity costs are outside numerator. This is a local scaling approximation for ordinary marginal support, not a claim one donor must finance two budgets or that effects scale linearly indefinitely.

g=.998 is an explicit judgment allowing territorial/nonresident effects, informed by USDA participant geography but not equal to an observed outcome-weighted share. 50 states plus DC residents only; noncitizen residents count. No headquarters-based apportionment.

q=(life/exposure)*utility*transport/(1+discount)^lag; PY=N*z*T*p*a*b; Q_all=PY*q-b*h; Q_USA=g*Q_all; price=10*C/Q_USA when Q_USA>0. C is two annual budgets; only one two-year continuity cohort, no perpetual policy credit. lag includes the delay to realized lifetime gains.

C
42995176 USD two-year recipient cost (judgment). Judgment extrapolation to two years at the observed annual recipient expense: 2 × (2024 Form 990 Part IX 21,482,914 + event direct costs 14,674); continuation at this scale is a model judgment. [frac-99024]
N
4000000 early-child participants per year (judgment). Round down 2023 4.464 million preschool count to 4 million prospective scale; not a forecast of 2027 enrollment or number at risk. [frac-snapdata]
z
0.05 additional benefit-year fraction conditional on relief (judgment). 5% annual continuity equivalent: e.g. 20% avoid a three-month gap. Net of state replacement and other causes; no observed FRAC-specific estimate. [frac-snap26]
T
2 years of changed continuity (judgment). Two years; no permanent repeal or lifelong benefit eligibility assumed. [frac-snap26]
p
0.1 coalition-induced probability increment (judgment). 10 percentage points for useful relief/implementation due to advocacy rather than political forces alone; weak prior, test zero. [frac-snap26]
a
0.05 FRAC share of incremental coalition contribution (judgment). 5% allows many national/state advocates and officials; not an empirical pivotality estimate.
b
0.5 marginal funding additionality (judgment). Half of normalized activity additional after other donations, public funds and reserve substitution. Separate from policy counterfactual. [frac-audit24] [frac-99024]
life
1.2 life-years for full early exposure (observed). Study's modeled treated life-expectancy estimate, not measured FRAC outcome or observed complete lifespans. [frac-childhealth]
exposure
5.75 years from conception to age five (observed). 5.75-year exposure window; linear division is an additional model assumption, not demonstrated effect of isolated gaps. [frac-childhealth]
utility
0.7 QALY weight per added life-year (judgment). 0.7 for later-life health, not study-measured utility.
transport
0.25 fraction of historical per-year effect retained (judgment). 25% discounts modern nutrition, short discontinuities, selection and lifetime extrapolation. Zero remains credible. [frac-childhealth] [frac-prenatal26]
discount
0.03 annual health discount rate (judgment). 3% standard modeling choice, not empirical finding.
lag
60 years until added life-year benefit (judgment). 60-year lumped approximation including prospective program delay; test 40–70 years rather than claim age-specific survival curve.
g
0.998 USA resident share (judgment). 99.8% permits territorial effects; uncertainty in relative health yield remains. [frac-snapdata]
h
0 other net QALY harm before funding additionality (judgment). Zero for unpriced portfolio centrally; positive h captures displacement, implementation harm or other downside independent of policy success.
WIC marginal enrollment
null additional retained participants per unrestricted dollar (unknown). Awards and participation totals do not identify causal incremental uptake or next-dollar allocation. [frac-wic25]

Weak-confidence partial-pathway best estimate: Cost: $43.0M; USA QALYs: 6.186549740111431; all-population QALYs: 6.198947635382195. {"C":42995176,"N":4000000,"z":0.05,"T":2,"p":0.1,"a":0.05,"b":0.5,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.25,"discount":0.03,"lag":60,"g":0.998,"h":0}; net expected additional child-years=1000; q=0.006198947635382195 discounted QALYs per additional child-year.

Stronger policy leverage and transport: Cost: $43.0M; USA QALYs: 643.5991865926889; all-population QALYs: 644.8889645217323. {"C":42995176,"N":4000000,"z":0.15,"T":2,"p":0.3,"a":0.1,"b":0.8,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.5,"discount":0.03,"lag":40,"g":0.998,"h":0}; net expected additional child-years=28800; q=0.022391977934782373 discounted QALYs per additional child-year.

Minimal avoidable churn and policy additionality: Cost: $43.0M; USA QALYs: 0.000736539842531653; all-population QALYs: 0.0007380158742802134. {"C":42995176,"N":4000000,"z":0.005,"T":2,"p":0.02,"a":0.01,"b":0.1,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.05,"discount":0.03,"lag":70,"g":0.998,"h":0}; net expected additional child-years=0.8; q=0.0009225198428502667 discounted QALYs per additional child-year.

Coalition changes no policy or implementation: Cost: $43.0M; USA QALYs: 0; all-population QALYs: 0. {"C":42995176,"N":4000000,"z":0.05,"T":2,"p":0,"a":0.05,"b":0.5,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.25,"discount":0.03,"lag":60,"g":0.998,"h":0}; net expected additional child-years=0; q=0.006198947635382195 discounted QALYs per additional child-year.

Historical health effect does not transfer: Cost: $43.0M; USA QALYs: 0; all-population QALYs: 0. {"C":42995176,"N":4000000,"z":0.05,"T":2,"p":0.1,"a":0.05,"b":0.5,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0,"discount":0.03,"lag":60,"g":0.998,"h":0}; net expected additional child-years=1000; q=0 discounted QALYs per additional child-year.

No continuity gain and other net harm: Cost: $43.0M; USA QALYs: -9.98; all-population QALYs: -10. {"C":42995176,"N":4000000,"z":0.05,"T":2,"p":0,"a":0.05,"b":0.5,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.25,"discount":0.03,"lag":60,"g":0.998,"h":20}; net expected additional child-years=0; q=0.006198947635382195 discounted QALYs per additional child-year.

Central continuity plus net harm elsewhere: Cost: $43.0M; USA QALYs: -3.793450259888569; all-population QALYs: -3.8010523646178047. {"C":42995176,"N":4000000,"z":0.05,"T":2,"p":0.1,"a":0.05,"b":0.5,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.25,"discount":0.03,"lag":60,"g":0.998,"h":20}; net expected additional child-years=1000; q=0.006198947635382195 discounted QALYs per additional child-year.

Most donations replace other funding: Cost: $43.0M; USA QALYs: 1.2373099480222862; all-population QALYs: 1.2397895270764392. {"C":42995176,"N":4000000,"z":0.05,"T":2,"p":0.1,"a":0.05,"b":0.1,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.25,"discount":0.03,"lag":60,"g":0.998,"h":0}; net expected additional child-years=200; q=0.006198947635382195 discounted QALYs per additional child-year.

Historical mean cost sensitivity: Cost: $30.7M; USA QALYs: 6.186549740111431; all-population QALYs: 6.198947635382195. {"C":30693129.333333332,"N":4000000,"z":0.05,"T":2,"p":0.1,"a":0.05,"b":0.5,"life":1.2,"exposure":5.75,"utility":0.7,"transport":0.25,"discount":0.03,"lag":60,"g":0.998,"h":0}; net expected additional child-years=1000; q=0.006198947635382195 discounted QALYs per additional child-year.

Counterfactual: z is additional continuity conditional on relief relative to states replacing funds or fixing administration anyway; p is advocacy coalition-induced relief/implementation probability increment, not total enactment probability. a apportions that increment to FRAC; b separately discounts donor replacement/reserve substitution. Only additional early-child receipt counted.

Attribution: Subjective policy branch tied to current administrative-funding advocacy, not credit for all national SNAP spending or all historic wins. FRAC's WIC and other portfolio effects are net zero centrally but unestimated, not established absent. Other coalition/public inputs remain necessary.

Very weak-confidence best estimate, not calibrated probabilities or confidence interval. Core policy and marginal enrollment quantities are judgments; historical clinical transport can be zero. Favorable/poor scenarios deliberately vary several parameters together. Full costs do not prove this partial model conservative if excluded effects are harmful.

Sensitivity

  • Central q=0.006198947635382195; about 1,000 expected additional child-years before geographic apportionment, not 400,000 FRAC-attributable child-years.
  • At q=0.006198947635382195 and g=.998, beating $1 million/$100 thousand per 10 USA QALYs requires 69497.82642372415 / 694978.2642372416 additional child-years after all attribution and funding adjustments.
  • Holding other central inputs, p*a must exceed 0.34748913211862076 for $1 million per 10 or 3.4748913211862074 for $100 thousand per 10, versus .005 central; the latter exceeds one. Stronger health/continuity or other pathways are needed.
  • Other net harm h above 12.39789527076439 QALYs before b erases central benefit. Harm is not multiplied by policy-success probability.
  • Every halving of z, p, a, b or transport doubles price absent other effects. Reducing the 60-year lag to 40 raises the present health value about 1.81-fold.
  • Using the historical mean reduces cost by about 29%, but the federal regrant portfolio is changing and neither historical cost base identifies next-dollar efficiency.
  • WIC could materially improve the assessment: require retained additional pregnancy/child episodes × causal QALYs per episode, net of alternative outreach, alongside a specific unrestricted-funding pathway. Do not divide national WIC health by FRAC's grant dollars.

Unresolved inputs

  • 2026–2028 unrestricted spending plan and marginal donor/public replacement, including WIC project continuation after its final award round.
  • Counterfactual effect of administrative relief on preschool benefit months and the coalition/FRAC incremental policy contribution.
  • Contemporary health utility and mortality effects of short benefit interruptions, with lag and persistence.
  • Causal WIC enrollment/retention by participant type, school-meal and adult pathways, public-resource opportunity costs and net harms.

5. Funding and previous grants

Recent financial scale is heavily influenced by federal regranting. The recipient return shows $14.27 million in government grants in 2024; year-end unrestricted net assets were about $0.81 million, while most net assets were restricted. Receivables and cash cannot be treated as unrestricted room for more funding. Return The consolidated audit describes a roughly $1.6 million reserve draw during 2024 and substantial grants payable; these are group context, not recipient-only free-cash figures. Audit

The latest disclosed finance year on the current index is 2024. No verified 2026 marginal hiring, implementation or matching-funds plan was found in the reviewed material. A donation could support urgent flexible policy work, replace other contributions, rebuild reserves, or support restricted-project overhead. The 50% additionality judgment is intentionally revisable and does not claim a specific unfunded opportunity. Finance index

The final application FAQ anticipated a January 2025–December 2026 project period and explicitly no further rounds. That is evidence of a planned end, not proof that current work has stopped or that federal support was terminated. Sustaining effective methods afterward could be valuable, but requires evaluation and a funded-versus-unfunded continuation plan. Round 2 FAQ

The verified official giving page names Food Research & Action Center as the recipient and links its EveryAction form. Use the general fund for this recipient-wide assessment, not a restricted initiative or the separate Action Council. The processor page identifies FRAC, but its JavaScript checkout and tax details were not independently inspected; the donation link therefore goes to FRAC's official giving page. Ways to donate

Annual expenses

Organization-level spending, including programs, administration and fundraising. The research list averages three comparable, consecutive full fiscal years when available.

  • FY 2022: $9.7M; Food Research & Action Center, EIN 23-7200739; excludes FRAC Action Council, 12-month period, Accrual Form 990 recipient expense plus direct event costs netted against revenue. Source
  • FY 2023: $14.8M; Food Research & Action Center, EIN 23-7200739; excludes FRAC Action Council, 12-month period, Accrual Form 990 recipient expense plus direct event costs netted against revenue. Source
  • FY 2024: $21.5M; Food Research & Action Center, EIN 23-7200739; excludes FRAC Action Council, 12-month period, Accrual Form 990 recipient expense plus direct event costs netted against revenue. Source

6. Sources

  1. 2024 Form 990, Parts I, III, VIII–XII. Food Research & Action Center. Published: not stated; retrieved: 2026-09-14.
  2. 2023 Form 990, Parts I, VIII and XII. Food Research & Action Center. Published: not stated; retrieved: 2026-09-14.
  3. 2022 Form 990, Parts I and VIII. Food Research & Action Center. Published: not stated; retrieved: 2026-09-14.
  4. 2024 consolidated financial statements and notes. Food Research & Action Center and FRAC Action Council. Published: not stated; retrieved: 2026-09-14.
  5. Financial disclosure index, latest listed year 2024. Food Research & Action Center. Published: not stated; retrieved: 2026-09-14.
  6. Second and final WIC CIAO subgrant round. FRAC / WIC CIAO. Published: 2025-01-07; retrieved: 2026-09-14.
  7. Why the SNAP Administrative Cost Shift Threatens State Budgets and Program Operations. Food Research & Action Center. Published: 2026-07-31; retrieved: 2026-09-14.
  8. SNAP: Changes in Federal-State Administrative Cost Sharing, proposed rule. USDA / Federal Register. Published: 2026-06-24; retrieved: 2026-09-14.
  9. Characteristics of SNAP Households: Fiscal Year 2023, Table B.14 and definitions. USDA / Mathematica. Published: not stated; retrieved: 2026-09-14.
  10. Is the Social Safety Net a Long-Term Investment? Large-Scale Evidence From the Food Stamps Program. Bailey, Hoynes, Rossin-Slater and Walker / Review of Economic Studies. Published: not stated; retrieved: 2026-09-14.
  11. How Does SNAP Access Prior to Pregnancy Affect Maternal and Infant Health Outcomes?. The Milbank Quarterly. Published: not stated; retrieved: 2026-09-14.
  12. Maternal and Child Outcomes Associated With WIC. AHRQ. Published: 2022-04-19; retrieved: 2026-09-14.
  13. WIC CIAO Round 2 final funding and evaluation FAQ. FRAC / WIC CIAO. Published: 2024-08-28; retrieved: 2026-09-14.
  14. Official giving options and recipient identification. Food Research & Action Center. Published: not stated; retrieved: 2026-09-14.
  15. Officially linked EveryAction donation form, recipient name visible; checkout not inspected. Food Research & Action Center / EveryAction. Published: not stated; retrieved: 2026-09-14.