The resulting instrumental variables (2SLS) models suggest that Medicaid coverage significantly increases the utilization of medical care by low-income children. Specifically, Medicaid is found to substantially decrease the probability of going without a visit to a doctor’s office and significantly increase probability of hospitalization in the previous year. Increased Medicaid coverage is also associated with a significantly higher probability of going to a doctor’s office than going to ER or hospital clinics. However, the estimation results provide no support for the hypothesis that Medicaid improves the health of low income children.

The paper is here.