Effective and scalable HIV prevention for adolescents in sub-Saharan Africa is needed. Cash transfers can reduce HIV incidence through reducing risk behaviours. However, questions remain about their effectiveness within national poverty-alleviation programmes, and their effects on different behaviours in boys and girls. In this case-control study, we interviewed South African adolescents (aged 10-18 years) between 2009 and 2012. We randomly selected census areas in two urban and two rural districts in two provinces in South Africa, including all homes with a resident adolescent. We assessed household receipt of state-provided child-focused cash transfers, incidence in the past year and prevalence of transactional sex, age-disparate sex, unprotected sex, multiple partners, and sex while drunk or after taking drugs. We used logistic regression after propensity score matching to assess the effect of cash transfers on these risky sexual behaviours.