Summer EBT
Coverage Atlas

The Summer EBT program

Summer Electronic Benefit Transfer (Summer EBT) — branded SUN Bucks by USDA in 2024 — was made permanent by the Consolidated Appropriations Act of 2023 (Section 502) after its successful pandemic-era predecessor, Pandemic EBT (P-EBT). It gives families of eligible children a grocery-spending benefit to use during the summer months when the National School Lunch Program (NSLP) and School Breakfast Program (SBP) are not in session.

• Benefit amount: $120 per eligible child per summer (statutory, indexed annually to the Thrifty Food Plan).
• Delivery: loaded onto an EBT card; spendable at any SNAP-authorized retailer, including most grocery stores and many farmers markets.
• Administrator: state agencies (usually the SNAP agency or Department of Education) run the program under USDA FNS oversight.
• Opt-in: states choose whether to participate each plan year, following a federal notification deadline set by FNS.

Eligibility rules

A child is eligible for Summer EBT if they meet any of the following in participating states:

1. Certified for free or reduced-price school meals at a NSLP/SBP school (the dominant path).
2. A member of a household receiving SNAP, TANF cash assistance, or FDPIR.
3. A foster child, homeless, migrant, runaway, or enrolled in Head Start.
4. Living in a household with income at or below 185% of the federal poverty line (the "income test"), documented via a short state application.

The 185% FPL bar is identical to the NSLP reduced-price meal cutoff. A child attending a Community Eligibility Provision (CEP) school is not automatically eligible — the child must still be individually certified via one of the routes above.

State participation (2025 plan year)

State-level participation status is coded by PantryPath from the USDA FNS SUN Bucks plan-approval list and state-agency press releases as of April 2026. Each state receives one of three enrollment-method tags:

• Automatic (auto-enroll) — the state identifies eligible children from existing records (NSLP certification, SNAP, TANF, Medicaid, foster) and issues SUN Bucks without requiring an application. Most participating states use this model.
• Hybrid — automatic enrollment for children already in a qualifying program (SNAP, NSLP certification, etc.), but the state requires a short application for children whose families are income-eligible but not already enrolled. This reaches fewer kids in practice but is easier for state agencies to stand up.
• None (opted out) — the state did not submit a 2025 plan to FNS, or explicitly declined to participate. Children in these states do not receive SUN Bucks benefits; the federal dollars attached to those kids are not drawn down.

Territories (Puerto Rico, Guam, U.S. Virgin Islands, American Samoa, Northern Mariana Islands) and federally-recognized Tribal nations participating under Indian Tribal Organization (ITO) plans are covered under separate SUN Bucks authorities and are not displayed at the state level in this atlas — they will appear in a Tribal addendum in a future refresh.

Data sources

All numeric sources are federal, publicly available, and do not require an API key. Citations with access dates live in the Sources section below.

Eligibility estimate — the 2.2 scaling factor

SAIPE publishes school-age children (5–17) in poverty — children in families with income at or below 100% of the federal poverty line — for every county annually. Summer EBT eligibility, however, uses a 185% FPL cutoff (the NSLP reduced-price threshold). The two thresholds differ by a large and stable multiplier.

To bridge them, we compute a national ratio from ACS table B17024 (Age by Ratio of Income to Poverty Level), which publishes counts of children at each poverty band. Using the 2023 ACS 1-year release:

eligible_children_est = saipe_children_5_17_in_poverty × 2.2

The 2.2 factor approximates the national ratio of children ≤185% FPL to children ≤100% FPL. It is applied uniformly across counties because county-level B17024 is suppressed for most counties due to sample size. Using a national ratio means counties with atypical income distributions (very deep poverty or a compressed distribution around the poverty line) are estimated less accurately than the national average. We surface this as a visualization aid, not a direct measurement — consistent with the standard disclaimer on all PantryPath derived indices.

The eligibility share metric is:

eligibility_share = eligible_children_est / saipe_children_5_17_universe

This ratio — the share of school-age children estimated eligible for Summer EBT — has the virtue of being monotone in poverty intensity and largely immune to the 2.2 factor (both numerator and denominator scale together on the 5–17 base). Use the share to compare county-to-county relative need; use the raw eligible count for absolute dollar impact.

Dollars flowing and dollars forfeited

The statutory FY2025 per-child benefit is $120 for the summer. We compute two county-level dollar figures:

• Dollars flowing — for counties in participating states: eligible_children_est × $120. Zero for counties in opt-out states.
• Dollars forfeited — for counties in opted-out states: eligible_children_est × $120. Zero for counties in participating states.

"Dollars forfeited" treats every eligible child in an opt-out state as receiving zero federal Summer EBT benefit. In reality, some of those kids may be reached by overlapping programs (SFSP meal sites, TEFAP commodities, philanthropy). But those programs pre-existed Summer EBT and are not a substitute for the $120-per-child grocery card that federal law offered the state. The framing mirrors USDA's own approach in its 2024 implementation report.

We do not estimate a "take-up rate" at the county level. USDA's implementation report shows that automatic-enrollment states reach 80–95% of eligible children in year 1, while hybrid states reach 40–60% depending on application outreach. But take-up is reported at the state level, not the county level, and applying a state average uniformly would hide real intra-state variation. Future refreshes may incorporate state-agency take-up data once FNS publishes it for the 2025 plan year.

State and national rollups

• State aggregates sum county-level counts and dollar figures. Eligibility share is re-derived at the state level from state-level SAIPE counts, not averaged from counties.
• National aggregates sum state-level values. The total eligible-children estimate is the sum across counties; national eligibility share is the national eligible count divided by the national 5–17 universe.
• Rankings surface six cross-state stories: most eligible kids, largest dollars forfeited, largest dollars flowing, highest eligibility share, automatic-enrollment leaders (by eligible-child volume), and the full opt-out list.

Limitations

• National scaling factor. The 2.2 multiplier is a national average. Counties with deeper poverty or unusual income distributions may be under- or over-estimated. Eligibility share (share of 5–17 population) is the more robust cross-county comparison metric.
• Participation status has a vintage. State decisions for the 2025 plan year were coded in April 2026. If a state reverses its decision mid-year the atlas will not reflect that until the next data refresh.
• Enrollment method is coded, not measured. The automatic / hybrid distinction is assigned from state plan documents; the line between "automatic for SNAP/NSLP-certified, application for income-only" (hybrid) and "automatic for everything" (automatic) is blurry in a few states.
• No take-up estimate. Dollar figures assume 100% take-up among eligible children. Actual issuance will be lower — especially in hybrid states — but no county-level take-up data exists yet for 2025.
• Territories and Tribes excluded. The 51-state view reflects 50 states + DC. Puerto Rico and the territories, plus ITO plans, follow separate USDA authorities and are not mapped here.
• SAIPE vintage. County estimates use 2023 SAIPE; income changes and demographic shifts through 2024 and early 2025 are not reflected.

Reproducibility

The entire atlas is driven by one pipeline script: scripts/summer-ebt/build-counties.mjs. It (1) fetches SAIPE 2023 for every county and state, (2) joins against the hardcoded STATE_PARTICIPATION_2025 table, (3) applies the 2.2 scaling factor and the $120 benefit, and (4) writes per-state + per-county JSON plus a national summary + rankings file. Emitted files live in src/data/summer-ebt/.

Pipeline metadata — dataset vintage, build timestamp, scaling factor, per-child benefit — is embedded in every emitted JSON file and surfaced on the atlas's index page. Annual refresh cadence: re-run the script each spring after FNS publishes the new plan year's participation list.