SUPPLEMENT TO “ INSURER COMPETITION IN HEALTH CARE MARKETS ” ( Econometrica

We estimate this model via maximum likelihood using our admission data. In each market, we normalize one hospital fixed effect to zero. We choose the largest hospital in each market to ensure comparability across markets. We define five diagnosis categories using ICD-9-CM codes and major diagnosis category (MDC) codes, as shown in Table A.I. The categories are cardiac, cancer, labor, digestive diseases, and neurological diseases. The sixth category, “other diagnoses,” includes all other categories in the data other than newborn babies (defined as events with MDC 15 where the patient is less than 5 years old). The hospital “service” variables are defined using American Hospital Association data for 2003–2004 (if observations are missing for a particular hospital in one year, we fill them in from the other). These variables summarize the services offered by each hospital; they cover cardiac, imaging, cancer, and birth services. Each hospital is rated on a scale from 0 to 1, where 1 implies that the hospital offers the least common of a list of relevant services and 0 implies that it offers none of the services. Details are given in Table A.II. Finally, since we do not observe household income for non-state agency enrollees (and we estimate our demand system using observed admissions from all enrollees), we use the mean household income in each zip code from Census data (winsorized at 5%).