Working Paper: NBER ID: w17937
Authors: Laurence Ales; Roozbeh Hosseini; Larry E. Jones
Abstract: In this paper we study the efficient allocation of health resources across individuals. We focus on the relation between health resources and income (taken as a proxy for productivity). In particular we determine the efficient level of the health care social safety net for the indigent. We assume that individuals have different life cycle profiles of productivity. Health care increases survival probability. We adopt the classical approach of welfare economics by considering how a central planner with an egalitarian (ex-ante) perspective would allocate resources. We show that, under the efficient allocation, health care spending increases with labor productivity, but only during the working years. Post retirement, everyone would get the same health care. Quantitatively, we find that the amount of inequality across the income distribution in the data is larger that what would be justified solely on the basis of production efficiency, but not drastically so. As a rough summary, in U.S. data top to bottom spending ratios are about 1.5 for most of the life cycle. Efficiency implies a decline from about 2 (at age 25) to 1 at retirement. We find larger inefficiencies in the lower part of the income distribution and in post retirement ages.
Keywords: Health Spending; Income Inequality; Welfare Economics
JEL Codes: H4; H51; I18; I38
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| labor productivity (J24) | health care spending (H51) |
| retirement (J26) | uniform allocation of health care (I14) |
| age (J14) | ratio of spending on top to bottom income quartiles (D31) |
| income quartiles (D31) | health care spending (H51) |
| age 55-65 (J26) | health care spending for bottom quartile (I18) |
| observed inequality in health spending (I14) | theoretical efficiency (D61) |
| current allocation of health care (H51) | level of social insurance (H55) |