Essays on health economics and risk preferences

Abstract

This dissertation is a collection of three essays on hospital response to regulation and risk preferences. Chapter 1 analyzes the Medicare Flex Program which allowed rural hospitals with fewer than 25 beds to convert to "Critical Access Hospital" status and receive cost-based Medicare reimbursement. Converting hospitals decrease their inpatient capacity by 30 percent on average to satisfy the bed requirement. They also drop services such as obstetrics, intensive care, and inpatient and outpatient surgery. Medicaid days fall by almost 30 percent at converting hospitals. The results suggest a 6 percent increase in neonatal mortality for high-risk babies delivered by mothers residing in counties with a hospital conversion. There is no significant effect on neonatal mortality among all births. Chapter 2 (co-authored with Liran Einav, Amy Finkelstein and Mark Cullen) analyzes the extent to which individuals' choices over five employer-provided insurance coverage decisions and one 401(k) investment decision exhibit systematic patterns, as would be implied by a general utility component of risk preferences. We provide evidence consistent with an important domain-general component that operates across all insurance choices. We find a considerably weaker relationship between one's insurance decisions and 401(k) asset allocation, although this relationship appears larger for more "financially sophisticated" individuals. Estimates from a stylized coverage choice model suggest that up to thirty percent of our sample makes choices that may be consistent across all six domains. Chapter 3 analyzes the effect of a California regulation mandating maximum patient-to-nurse ratios for inpatient hospital units. I compare changes in inputs for hospitals for which the legislation was more or less binding, based on the initial patient-to-nurse ratio. Hospitals with higher baseline ratios decrease patient-to-nurse ratios by substituting toward licensed nurses, rather than decreasing patient days or lengths of stay. There is suggestive evidence of substitution away from aides and orderlies. The skill ratio of licensed nurses decreases slightly as hospitals hire relatively more licensed vocational nurses. Total costs increase in hospitals with higher baseline ratios. These results are slightly smaller and no longer significant in robustness checks. Estimating these effects on a sample of unaffected units generally finds small, insignificant effects.