Do Provider Networks Impact Health Care Use and Spending? Evidence from Random Assignment in Medicaid Managed Care

A key feature of health insurance plans in the United States is the breadth of their provider networks (i.e. the number of providers they contract with). To study the impact of network breadth, I leverage a natural experiment in New York where Medicaid recipients are randomly assigned to private health plans that contract with different sets of physicians and hospitals. To measure network breadth, I use estimates from a structural model of demand for physicians and hospitals. Combining these measures with administrative health records for over 50,000 randomly assigned recipients, I find that assignment to a narrower network reduces the use of health care services and spending at the expense of consumer satisfaction, with little impact on traditional measures of access and quality. The spending reductions came primarily from outpatient specialty care and prescription drugs, and were driven by narrower physician network breadth. The reductions in consumer satisfaction, on the other hand, were driven by narrower hospital network breadth and fully mediated by whether a consumers’ usual source of care (often a hospital for Medicaid recipients) was covered by their assigned plan. Taken together, the results suggest that alternative auto-assignment policies could improve the tradeoff between spending and satisfaction.