Estimates of Crowd-Out from a Public Health Insurance Expansion Using Administrative Data

We use administrative data from Wisconsin to estimate the percent of individuals newly enrolled in public health coverage that had access to private, employer-sponsored health insurance at the time of their enrollment and the percent that was uninsured. We also estimate the percent of new enrollees that dropped private coverage in the six months prior to or following enrollment. We estimate that, among all new enrollees and after expansion of eligibility for public coverage, approximately 21% had access to private health insurance at the time of enrollment and that only 10% dropped this coverage. We also estimate strict bounds for these percentages and find that the percentage of new enrollees with private insurance coverage at the time of enrollment lies between 16% and 29% and the percentage that dropped private coverage in favor of public insurance lies between 4% and 18%. These estimates of crowd-out are relatively low compared with estimates from the literature based on Medicaid and Children’s Health Insurance Program expansions, although based both on different data and on a different method. In order to facilitate comparison of our method with previous studies, we develop a conceptual method for understanding the various types of crowd-out measures.