Working Paper: NBER ID: w28864
Authors: Barton Hamilton; Andrés Hincapié; Emma C. Kalish; Nicholas W. Papageorge
Abstract: Health-maximizing and welfare-maximizing behaviors can be at odds, especially among disadvantaged groups, which can contribute to health disparities. To investigate this point, we estimate a lifecycle model of medication and labor supply decisions using data on HIV-positive men. We use the model to evaluate the disparate consequences of an effective HIV treatment innovation that had harsh side effects: HAART. Measured in lifetime utility gains, HAART disproportionately benefited patients with more education. Lower-educated men were more likely to avoid HAART due to its side effects that interfered with work. To illustrate the wedge between health and welfare, we simulate the effects of a HAART treatment mandate, which mimics assignment to treatment in a clinical trial. The mandate improves health, which would be viewed as a success in a randomized trial. However, clinical trials, which often focus solely on health outcomes, can mask downsides of the treatment including its distributional consequences: the mandate increases inequality as measured by lifetime welfare because lower-educated men are more likely to stop working due to HAART-induced side effects. In contrast, a counterfactual policy simulation that provides a non-labor income subsidy increases HAART adoption and improves health, especially among lower-education individuals. Broadly, our study illustrates that the evaluation of medical innovations may be incomplete absent an understanding of their distributional consequences across different groups of patients.
Keywords: health disparities; HIV treatment; HAART; medical innovation; labor supply; welfare
JEL Codes: I12; I14; I20; J2; O31
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| HAART (Q16) | health-maximizing behaviors (I12) |
| health-maximizing behaviors (I12) | welfare-maximizing behaviors (D69) |
| HAART (Q16) | welfare-maximizing behaviors (D69) |
| lower education (I24) | HAART adoption (F35) |
| HAART side effects (I12) | employment decisions (M51) |
| HAART treatment mandate (I18) | health outcomes (I14) |
| HAART treatment mandate (I18) | welfare inequality (I30) |
| non-labor income subsidies (J32) | HAART adoption (F35) |
| non-labor income subsidies (J32) | health among lower-educated individuals (I14) |
| medical innovations (O35) | health disparities (I14) |