This study examines whether privately insured patients avoid hospitals with large Medicaid populations. I use a conditional logit model of hospital choice to determine whether the size of a hospital's Medicaid population affects the probability that a privately insured patient will choose that hospital. I focus on the metropolitan area of Tampa, Florida, in the years 1994–1996. I control for hospital fixed effects, hospital-specific time trends, patients' driving time to the hospital, and interactions between patient and hospital characteristics. I also instrument for the Medicaid population using the predicted Medicaid population. The results show that privately insured patients are less likely to choose a hospital if it served a larger number of Medicaid patients who were admitted through the emergency department in the previous 6 months. The effect persists over time—an additional 6-month lag cuts the effect in half. Capacity constraints do not seem to be the reason for the effect. I show that the Medicaid effect size could have a moderate effect on the profits of some hospitals. Although limited in scope, this study suggests that hospitals may experience a negative effect on their private revenues when they admit a large population of Medicaid patients. Copyright © 2012 John Wiley & Sons, Ltd.