What percentage of SNAP payments are due to trafficking versus recipient error?

Executive summary — what the evidence actually shows about SNAP trafficking versus recipient error

The most consistent, recent federal estimate places SNAP trafficking at roughly 1.5–2.0 percent of benefits during 2015–2017, equal to about $1.08–$1.27 billion annually by that period’s redemption volume, while national improper payment measures show overall error and fraud in SNAP in the single-digit-to-low-double-digit range (roughly 6.8–11.7 percent in recent years), but those measures do not parse trafficking from recipient error ^{[1] [2] [3] [4]}. A private-sector study published in June 2025 estimates a different breakdown of fraud-related losses — about 20 percent attributable to trafficking and about 38 percent to inadvertent household (recipient) errors — but that study uses different methods and scopes than USDA federal trafficking estimates, so the figures are not directly comparable ^[5]. Overall, federal data show trafficking is a small fraction of total benefits, while recipient error and broader improper payments account for a substantially larger share of measured inaccuracies, though exact relative percentages depend on methodology and definitions ^{[6] [4]}.

1. Why the trafficking headline (1.5–2.0%) matters but doesn’t settle the trafficking-vs-recipient-error question

Federal analyses measuring trafficking produced a point estimate near 1.6 percent for the 2015–2017 window, with a range presented up to about 2.0 percent and associated dollar estimates of roughly $1.08 to $1.27 billion ^{[1] [2]}. Those trafficking estimates come from retailer-focused investigations and models that quantify benefits redeemed in ways consistent with trafficking patterns, with small stores accounting for most trafficked redemptions and store violation rates in the low teens, underscoring retail-side enforcement issues ^[2]. However, federal trafficking studies purposely target retailer trafficking and use legal/transactional markers distinct from household eligibility or reporting errors; therefore, the trafficking rate cannot be algebraically subtracted from overall improper payment rates to produce a clean trafficking-versus-recipient-error split without additional, harmonized measurement work ^{[1] [2]}.

2. The broader improper payment picture: big numbers, mixed causes

USDA’s reported national payment error rates and government oversight findings show substantially higher aggregates of improper payments than trafficking alone, with recent fiscal-year reporting indicating single-digit to low double-digit improper payment rates — for example, a national payment error rate reported for FY2024 and an 11.7 percent figure cited for FY2023 improper payments in oversight reviews ^{[6] [4]}. The National Payment Error Rate (NPER) aggregates multiple causes, including administrative mistakes, eligibility calculation errors, inadvertent recipient mistakes, and some fraud, but it does not separately identify trafficking as a distinct share, so NPER cannot be used alone to quantify recipient error versus trafficking ^{[3] [4]}. Government audits and GAO reviews have repeatedly called for better segmentation of error types to support targeted remedies, highlighting a measurement gap ^[4].

3. Private-sector estimates complicate the picture — different methods, different results

A LexisNexis private-sector study published in June 2025 reports that roughly 20 percent of SNAP fraud losses are from trafficking and about 38 percent from inadvertent household errors, suggesting a larger relative role for recipient mistakes within the study’s fraud-loss framing ^[5]. That study uses a different fraud-loss taxonomy and data sources than USDA analyses, focusing on fraud-loss attribution rather than the programwide trafficking rate derived from retailer redemption analyses. Because the LexisNexis study counts “fraud losses” and apportions causes within that narrower bucket, its percentages should not be conflated with the federal trafficking rate expressed as a share of total benefits without careful reconciliation of definitions and denominators ^[5].

4. Where methodological differences drive divergent messages

Three core methodological splits explain why a single, authoritative trafficking-versus-recipient-error percentage does not exist: first, denominator choice (total benefits vs. measured improper payments vs. fraud losses) changes the magnitude; second, unit of analysis (retail redemptions vs. household eligibility records vs. combined administrative data) yields different signals; third, scope and attribution rules (what counts as trafficking, what counts as inadvertent error, how mixed incidents are coded) differ between USDA, oversight bodies, and private studies ^{[1] [2] [4] [5]}. These methodological gaps are well-documented by federal oversight recommendations calling for improved trafficking estimates and clearer breakdowns in program integrity reporting ^[4].

5. Bottom line with policy context: trafficking small but not negligible; recipient error and system flaws dominate measured impropriety

Federal trafficking estimates show trafficking is a relatively small share of total SNAP benefits (about 1.5–2.0 percent in the referenced federal window), while aggregate improper payments are much larger and include a sizable component that is not trafficking but stems from recipient error and administrative problems; private-sector studies attribute a larger share of fraud losses to recipient mistakes, but those figures are not interchangeable with federal rates without harmonized methods ^{[1] [2] [4] [5]}. Policymakers and program managers should therefore treat trafficking as a focused enforcement problem concentrated at the retail end, while recipient error and state administrative processes remain the principal drivers of measured improper payments, warranting broad process-improvement, accuracy, and anti-fraud investments informed by better, standardized measurement ^{[4] [2]}.