Estimating Behavioral Response to the AIDS Epidemic

Abstract

AbstractThe elasticity of risky sexual behavior to changes in local HIV infection prevalence is estimated using a longitudinal survey of the sexual behavior and health of gay men in San Francisco during the 1980s. An average respondent decreases risky behavior by about 5% in response to a 10% increase in disease prevalence. The average response obscures substantial variation across respondents: High-risk people reduce risky behavior less than low-risk people as prevalence increases. This result is consistent with the predictions of theoretical economic epidemiology and has implications for epidemic dynamics.