The South African Government's anti-poverty strategy has been largely dominated by unconditional cash transfers. A growing body of literature examines the impacts of these transfers on a range of socio-economic outcomes; however there is little discussion of why such impacts are important. Without an explicit conceptual framework within which to examine these effects, evaluating their likely long term poverty impacts remains problematic. The focus of this research is to distil the current thinking on poverty and social protection to establish an appropriate theoretical framework within which to appraise anti-poverty measures. 'An Assets-Augmented' Capabilities Framework is proposed with a focus on asset-building as the primary means of poverty reduction. Focusing on the Child Support Grant (CSG), empirical analysis is then applied to examine whether the underlying causes of poverty are being addressed. Using the National Income Dynamics Study (NIDS) Wave 1 Dataset, the effects of the CSG on households' expenditure patterns are examined to assess the extent to which CSG-recipients invest in asset-building. A propensity score matching method is employed to construct an appropriate counterfactual. Households receiving the CSG are not found to exhibit significantly different expenditure patterns compared to a control group and cannot therefore be argued to invest differently in assets. This key finding provides evidence that the CSG primarily addresses the symptoms of poverty and cannot be expected to generate sustainable poverty reduction. A linking approach is thus proposed to combine the cash transfer element of the grant with more explicitly promotional measures in order to seek greater asset effects. A key recommendation is to ensure that the CSG acts as a gateway to other complementary services and benefits in order to increase the value of the grant with relatively little additional effort or cost.