There is a gap in the literature in understanding how cash transfer programmes affect mental health. We aim to fill this gap by conceptualising and estimating the mediation effects of an unconditional cash transfer programme on mental health. We use a sample of 4,535 adults living below the South African poverty line in four waves (2008–2014) of the South African National Income Dynamics Study. We use information on individual exposure to South Africa's largest unconditional cash transfer programme, the Child Support Grant. Mental health is measured by the 10-item version of the Centre for Epidemiological Depression Scale. We use the product of the coefficient method for the mediation analysis in combination with instrumental variable estimation. We find that physical health and lifestyle factors mediate the relationship of the unconditional cash transfer programme, each explaining about eight percent and 16% of the total positive effect. Our findings show that individuals living in poverty make investment decisions that are positive for their mental health, which has strong implications for policy makers.