Administrative Burden and Medicaid Coverage Loss: Evidence from a Field Experiment

Each year, millions of people lose access to safety-net programs simply because they do not complete required administrative steps. To study how these “transaction costs” affect Medicaid coverage, we conducted a large-scale field experiment that randomized outreach efforts offering personalized support during the renewal period. The intervention—providing both information and direct assistance—was designed to ease the renewal process without altering eligibility, so any observed effects reflect reductions in unnecessary coverage loss among eligible participants.

Pre-recorded calls offering free, one-on-one help from health insurance navigators raised Medicaid renewal rates by 1 percentage point (a 1.5% increase compared to the control group). Still, at least 20% of listed phone numbers were outdated, meaning many beneficiaries likely never received the call. Among those reached, the largest improvements occurred for tribal members (8 percentage points, or 13%) and children (3 percentage points, or 4%). Gains were also seen among lower-income individuals (2 percentage points, or 4%) and those with chronic illnesses (3 percentage points, or 4%).

Overall, the findings show that eligible individuals—especially those in need of care—often lose Medicaid because of burdensome administrative requirements. This challenges the common justification for enrollment barriers: that they selectively discourage participation by those who do not truly need or value the benefits.

Myerson, R. M., Espeseth, A., & Dague, L. (2025). Navigating Medicaid: Experimental evidence on administrative burden and coverage loss (NBER Working Paper No. 34191). National Bureau of Economic Research. https://doi.org/10.3386/w34191