Working paper number:
Many HIV testing programs in Africa and elsewhere aim to reduce risk-taking behaviors by providing individuals with information about their own HIV status. This paper examines how beliefs about own HIV status affect risky sexual behavior using data from married couples living in three regions of Malawi. Risky behavior is measured as the propensity to engage in extramarital affairs or to not use condoms. The empirical analysis is based on two panel surveys for years 2004 and 2006 from the Malawi Diffusion and Ideational Change Project (MDICP) and from an experimental HIV testing intervention carried out in 2004 that provided randomized incentives for picking up test results. Most individuals participating in the MDICP testing learned that they were HIV negative and a small fraction that they were positive. Controlling for potential endogeneity between beliefs and risk-taking, we find that downward revisions in the subjective belief of being HIV positive lead to decreases in the propensity to engage in extra-marital affairs but have no effect on condom use. These results are generally supported by survey questions that directly elicited from respondents how participating in testing altered their behavior. We show that the estimates provide a lower bound in the presence of measurement error in extra-marital affairs.
de Paula, Aureo, Gil Shapira, Petra E. Todd. 2008. "How Beliefs about HIV Status Affect Risky Behaviors: Evidence from Malawi." PARC Working Paper Series, WPS 08-09.