This paper considers the labour market effects of an unconditional cash transfer targeted at the unemployed in a context of extreme unemployment. Using a staggered, heterogeneity robust difference-in-differences design applied to panel labour force survey data, we estimate the contemporaneous and dynamic effects of a new transfer introduced in South Africa, the Social Relief of Distress grant, the first labour market-linked transfer in the country’s history. We find that, on aggregate, receipt has positive effects on the probabilities of job search, trying to start a business, and employment. The latter effects are driven by effects on wage and informal sector employment. We show that employment effects are positive for the unemployed who are either actively searching for work or trying to start a business, as well as for those who are not, but they are substantially larger for the former. This indicates that the transfer both encourages and improves the efficiency of labour market activity by addressing labour market constraints but highlights the importance of active labour market engagement for improving employment prospects through the transfer. However, these employment effects are only evident in the short-term and quickly become and remain null in the longer-term. These results suggest that cash transfers can help reduce labour market constraints but such gains need not translate into better longer-term employment prospects in high-unemployment contexts.