Website CV Paper
Fields
Health, Development, Behavioral & Experimental, Labor
Job market paper
Can a Temporary Information Shock Generate Lasting Demand? Evidence from a Condom Experiment in Ghana
Condoms protect against both HIV and unintended pregnancy, yet many young adults use them inconsistently. Information can shift the beliefs that shape health behavior, but belief changes do not always translate into behavior change, and they may not last. Using a randomized experiment with 2,151 male students in Ghana, I examine which belief changes translate into behavior change and whether a one-time information shock can generate lasting demand for condoms. Health risk information increases condom demand by 4.1 percentage points after delivery and by 7.1 percentage points eight weeks later, relative to a control rate of 13 percent. The effect is larger at eight weeks even though the shift in beliefs about condom safety is no longer detectable. This pattern is consistent with experiential learning, whereby initial use lowers barriers to later use. In contrast, peer information corrects beliefs about stigma around condom purchase but does not increase demand. Brief, low-cost health risk messages can thus generate persistent demand for preventive health products, whereas correcting beliefs about stigma may not suffice on its own.