Working Paper: NBER ID: w13049
Authors: Emily Oster
Abstract: The response of sexual behavior to HIV in Africa is an important input to predicting the path of the epidemic and to focusing prevention efforts. Existing estimates suggest limited behavioral response, but fail to take into account possible differences across individuals. A simple model of sexual behavior choice among forward-looking individuals implies that behavioral response should be larger for those with lower non-HIV mortality risks and those who are richer. I estimate behavioral response using a new instrumental variables strategy, instrumenting for HIV prevalence with distance to the origin of the virus. I find low response on average, consistent with existing literature, but larger responses for those who face lower non-HIV mortality and for those who are richer. I also show suggestive evidence, based on a very simple calibration, that the magnitude of behavioral response in Africa is of a similar order of magnitude to that among gay men in the United States, once differences in income and life expectancy are taken into account.
Keywords: HIV; Sexual Behavior; Africa; Instrumental Variables
JEL Codes: I10; J17; O12
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| HIV prevalence (O15) | risky sexual behavior (I12) |
| distance to the origin of the virus (R12) | HIV prevalence (O15) |
| HIV prevalence (O15) | sexual behavior (behavioral response varies by individual factors) (C91) |
| HIV prevalence (O15) | behavioral response in Africa (C92) |