Recovering Elasticities in Health Expenditure Using Nonlinear Reimbursement
(Job Market Paper) [PDF]
ABSTRACT: U.S. health care costs are rising rapidly. Several recent initiatives attribute this to the fact that patients do not bear the full cost of their care. Measures which decrease reimbursement of patients’ expenditures include high deductible plans, Health Savings Accounts (HSAs), and the Medicare Drug Plan "doughnut hole." But how does health expenditure change in response to decreasing reimbursement rates? To answer this question, this paper develops an estimation method which takes advantage of pervasive nonlinear reimbursement in health insurance contracts, including all the aforementioned reform initiatives. This estimation method avoids the bias present in the previous literature by addressing the problem that nonlinearities cause expenditure and reimbursement rates to be simultaneously determined. Using a unique claims-level dataset of employer-sponsored health insurance, I find a tight range of local elasticities between -0.25 and -0.33 on the range of average U.S. spending. Finally, I use my estimates to conduct a counterfactual policy experiment measuring the inefficiency of full reimbursement, and find evidence of welfare loss due to overconsumption. The method developed here can be used on many important policies with nonlinear reimbursement, such as Medicare Parts A and B, which previous tools could not address.
References
|
Professor Patrick Bajari |
Professor Robert Town |
|
Professor Thomas Holmes |
Professor Amil Petrin |