brain/
questionopenpolitics

Are MN and CA state-program-fraud rates actually outliers among states?

Notes

Are MN and CA state-program-fraud rates actually outliers among states?

One-line summary: MN and CA dominate the 2024–2026 state-program-fraud headlines, but standardized Medicaid Fraud Control Unit and per-capita-investigation data do not put either state at the top of fraud rankings. The headline-vs-data gap is itself the methodologically interesting question.

The question

Is the political narrative ("MN and CA are uniquely fraud-ridden") supported by per-capita or per-enrollee data, or is it a selection artifact of (a) reporting concentration, (b) prosecutorial visibility, and (c) partisan targeting? Specifically: are MN's per-capita rates of state-program fraud higher than (say) Florida's or Texas's, controlling for program scale?

Why it matters

This is the methodologically important question beneath the entire MN/CA fraud-narrative thread. If MN/CA are not outliers per capita, the headlines are a reporting and political-targeting artifact, and the appropriate federal-policy response is broad structural reform (per state-administered-federal-program-fraud-vulnerability) rather than state-specific punitive action (like trump-2026-childcare-funding-freeze). If they are outliers, state-specific action is more defensible.

What we currently believe

Evidence we have

  • Texas + CA + NY + FL + IL have the highest enrollment AND the highest case counts (Texas: 1,532 investigations; CA: 1,452; NY: 821; FL: 809; IL: 502). Per 2026-05-13-autoresearch-recent-fraud-minnesota-california-hospice-daycare.
  • Per-capita ranking changes the picture: Delaware, Hawaii, MO, AR top by per-enrollee investigation rates.
  • KFF caveat directly addresses the question: "Raw recovery dollars, conviction counts, per-enrollee investigation rates, and automated billing-risk flags each illuminate different facets of Medicaid fraud; readers should treat single-metric headlines with caution because measurement choices, legal settlements' structure, enforcement capacity, and political dynamics all shape the picture."
  • CMS expanded hospice-fraud oversight to CA, AZ, NV, TX, OH, GA, then FL — not purely partisan, but not consistent with "MN/CA are the uniquely-bad states" framing.
  • North Carolina autism therapy comparator (round 2 update): NC Medicaid autism therapy spending grew from $1.4M to $660M in five years (47,000% growth), with $1B projected by 2027. State Auditor Dave Boliek flagged the anomaly for investigation. Mechanism per Discern Money: "loosened standards, aggressive provider marketing, telehealth loopholes, and weak verification." The same article: "Similar spikes have drawn audits in multiple states." This is bigger than MN's autism scheme by an order of magnitude. The MN/CA framing of the state-program-fraud story is partial; the pattern is national.

Evidence we need

  • Per-capita / per-enrollee fraud rate by state, controlling for program mix.
  • Per-program comparison: how does the MN EIDBI fraud rate compare to similar programs in other states?
  • CMS Improper Payments breakdown by state (FY2025 fact sheet referenced in 2026-05-13-autoresearch-recent-fraud-minnesota-california-hospice-daycare but fetch timed out).
  • A comparable case study: if FL or TX had been investigated with House Oversight intensity, would the findings be comparable? Counterfactual is hard to establish but the question matters.

How to resolve

  • Track CMS FY2025 Improper Payments Fact Sheet for state-level breakdown.
  • Track Medicaid Fraud Control Unit annual reports for per-state rates.
  • Watch for any DOJ or GAO report that does cross-state fraud-rate comparisons with methodology.
  • Note as a hard-to-fully-resolve question — different metrics will continue to support different framings.

Related

Referenced by