Decisions with uncertain outcomes are often made by one party in settings where another party bears the consequences. Whenever an individual is delegated to make decisions that affect others, such as in the typical corporate structure, does the individual make decisions that reflect the risk preferences of the party bearing the consequences? We examine this question in two simple settings, lottery choices and sealed-bid auctions, using controlled laboratory experiments. We find that when an individual makes a decision for an anonymous stranger, there is a tendency to exhibit less risk aversion. This reduction in risk aversion is relative to his or her own preferences, and it is also relative to his or her belief about the preferences of others. This result has significant implications for the design of contracts between principals and agents.

Recursive expected utility, Risk, Ambiguity, C9, C44, C91,

The proposed measure provides an important foundation to improve APM prescribing practices without penalizing NHs when there are limited alternative treatments available.

Acknowledgements: We are greatly indebted to NBER for the permission of using the AHA Annual Survey, datasets. We would like to express our sincere thanks to Mohan Ramanujan and Jean Roth for their kind assistance. We would also like to thank Kristin Madison for access to her hospital system membership database. AbstractWe examine the evolving structure of the U.S. hospital industry since 1970, focusing on how ownership form influences entry and exit behavior. We develop theoretical predictions based on the model of Lakdawalla and Philipson, in which for-profit and notfor-profit hospitals differ regarding their objectives and costs of capital. The model predicts for-profits would be quicker to enter and exit than not-for-profits in response to changing market conditions. We test this hypothesis using data for all U.S. hospitals from 1984 through 2000. Examining annual and regional entry and exit rates, for-profit hospitals consistently have higher entry and exit rates than not-for-profits. Econometric modeling of entry and exit rates yields similar patterns. Estimates of an ordered probit model of entry indicate that entry is more responsive to demand changes for for-profit than not-for-profit hospitals. Estimates of a discrete hazard model for exit similarly indicate that negative demand shifts increase the probability of exit more for for-profits than not-for-profits. Finally, membership in a hospital chain significantly decreases the probability of exit for for-profits, but not not-for-profits.

A small but growing number of states are turning to accountable care concepts to improve their Medicaid programs. In 2011 New Jersey enacted the Medicaid Accountable Care Organization (ACO) Demonstration Project to offer local provider coalitions the opportunity to share any savings they generate. Impetus came from initiatives in Camden that aim to reduce costs through improved care coordination among hospital high users and that have received considerable media attention and substantial federal and private grant support. Though broadly similar to Medicare and commercial ACOs, the New Jersey demonstration addresses the unique concerns faced by Medicaid populations. Using hospital all-payer billing data, we estimate savings from care improvement efforts among inpatient and emergency department high users in thirteen communities that are candidates for participation in the New Jersey demonstration. We also examine their characteristics to inform Medicaid accountable care strategies. We find substantial variation in the share of high-user hospital patients across the study communities and high rates of avoidable use and costs among these patients. The potential savings among Medicaid enrollees are considerable, particularly if Medicaid ACOs can develop ways to successfully address the high burden of chronic illness and behavioral health conditions prevalent in the prospective demonstration communities.