Legislative auditors find Kansas at-risk funding may overcount students tied to free lunch program

Heidi, an audit presenter for the Legislative Division of Post Audit, told the Legislative Post Audit Committee that the number of students the state counts for at-risk funding "appears to be significantly more than the number of students who may be eligible for the free lunch program." The audit examined how free-lunch eligibility is determined and how that count is used to allocate state at-risk funding.

The audit matters because Kansas school districts received roughly $563 million in at-risk and high-density at-risk funding in the 2024 school year, and the state allocates much of that money based on counts of students eligible for the National School Lunch Program. The audit found evidence that the state likely overpaid at-risk funding in 2024 and identified limitations that make a full accuracy check difficult.

Legislative auditors reviewed records for a random sample of 769 students who were listed as eligible for free lunches in the 2024 school year and focused verification work on the subset who submitted household income applications. For 122 students whose families submitted National School Lunch Program applications, auditors matched application information to state Department of Revenue tax records or Department of Labor wage data when available. Of the 108 verifiable applications, auditors concluded 68 did not appear to meet the income eligibility thresholds the applications claimed. The report projects that, among students who qualified by application, between roughly 52% and 74% likely were ineligible, and when those sample results are projected to the statewide population auditors estimate the state may have overpaid approximately $38 million to $53 million in at-risk funding for the 2024 school year.

The report includes careful caveats. Auditors stressed the application is a snapshot in time while tax returns report annual income, and auditors said they could not always verify household size from outside records, both of which could cause mismatches between what an application showed and what tax or wage records later reflected. Audit staff also noted that the sample-based projection applies only to students who qualified by submitting an application, which auditors estimate represented about 16% of the free-lunch population in 2024.

The audit highlights three ways students become eligible for free lunch: direct certification (automatic eligibility through programs such as SNAP, TANF or certain Medicaid benefits, foster care, homelessness, migrant/runaway status), submission of a household income application, and schools participating in the USDA Community Eligibility Provision (which provides free lunch to all students at participating schools). Auditors observed that direct certification grew from about 46% of free-lunch students in school year 2021 to nearly 80% in 2024, leaving the majority of eligibility determinations in the hands of other state agencies that auditors could not fully verify.

Heidi told the committee that auditors could not verify the accuracy of direct certifications because those determinations rely on benefit eligibility decisions by agencies such as the Department for Children and Families (DCF) and the Kansas Department of Health and Environment (KDHE). "For 80% of the students who receive free lunch in 2024, we don't know if they were approved accurately or not," she said, noting that verifying those decisions was outside the scope of the audit. Auditors also described federal rules that limit districts' ability to verify National School Lunch applications: USDA rules generally require districts to verify only a small percentage of applications (3% of approvals or 3,000 applications, whichever is less), and state auditors said KSDE lacks authority and access to independently verify household incomes reported on applications.

The auditors reviewed household economic surveys that some students at Community Eligibility Provision schools submit for at-risk counts and found that those surveys accounted for only a small share of free-lunch students (about 4%). Of the 32 household economic surveys in the sample, auditors could verify only 16 with tax or wage records and concluded nine of the 16 did not appear to meet income eligibility for free lunch by income; however, because Community Eligibility schools provide free lunch to all students regardless of income, those students still receive free meals even if they are not counted for at-risk funding.

Auditors recommended the legislature consider whether basing at-risk funding on the free-lunch count remains appropriate, given program changes that have expanded the routes to free-lunch eligibility and the audit's evidence that many students counted by application may not have met the income thresholds used for at-risk funding. The audit noted that free-lunch eligibility was originally correlated with student poverty and academic risk but that changes such as expanded direct certification and community eligibility have broadened the population receiving free lunch.

Committee members asked several follow-up questions. Senators and representatives pressed audit staff about the limits of verification and whether KDHE or DCF could provide better data; Deputy Commissioner Frank Harwood of the Department of Education agreed to supply additional information about district verification results when available. Committee members also questioned whether USDA rules or state data-sharing restrictions (for example, Department of Revenue confidentiality) prevented more thorough verification outside the narrow frequency USDA permits.

The audit does not conclude that districts intentionally miscounted students nor that specific agencies committed errors; instead, it documents how the combination of federal program design, state agency processes and a large increase in direct certification make it difficult to judge the accuracy of the statewide free-lunch count and thus the at-risk funding allocation. The audit presentation and accompanying report are intended to inform legislative policy discussions about whether the at-risk funding formula should be revised or whether the state should seek ways to improve verification and interagency data sharing to reduce potential overpayments.