Abstract
This article examines whether substituting a health maintenance organization (HMO) for traditional fee-for-service (FFS) Medicaid insurance reduces the cost of children's health care. The estimation is complicated by the fact that in nonrandomized settings, unobserved selection can bias estimates of HMO performance. To control for selection, researchers often rely on parametric assumptions or instrumental variables estimation to compute selection-free estimates. But the robustness of these approaches has been questioned. We pursue a different approach based on semiparametric maximum likelihood techniques. Monte Carlo and applied economic studies have shown this method to be quite robust in a variety of contexts. We apply this model to data from a self-selected sample of children in either a Medicaid HMO or a traditional FFS in Florida. After controlling for selection, we estimate that the HMO reduced expenditures on children by 9.1%. Conversely, a model assuming no selection predicts no savings from the HMO. We also validate our estimates by comparing our results with those obtained from a randomized sample of HMO and FFS enrollees. These indicate that the HMO reduces expenditures by 13.6%. We conclude that selection can substantially bias estimates of HMO impact and that this technique provides a potentially useful method for accounting for this bias.