This minor dissertation unpacks the relationship between poverty and agriculture in South Africa. Agriculture has been identified as one of the key pillars of the South African economic development in the national development plan. The World Bank acknowledges that agriculture will continue to play a major role in reaching Millennium Development Goals. Therefore, this minor dissertation seeks to analyse the poverty determinants in South Africa and the agricultural linkages with the help of the National Income Dynamic Study (NIDS) panel data. The NIDS data set allows for an intertemporal comparison among the same households. The main aim of this study is to assess how agricultural activities such as subsistence farming are related to poverty in South Africa. The study hypothesizes that agriculture is a significant determinant of household poverty. The dynamics of poverty are revealed using the Foster, Greer and Thorbecke (FGT) poverty indices for the four waves of the NIDS data set. The study runs a dynamic random effect Probit model which assesses the link between poverty status, land ownership, crop growing, major crop failure and a host of other explanatory variables. Land ownership and major crop failure are found to be key determinants of the poverty status. The impact of demographics on poverty status is consistent with other poverty studies. People in rural areas generally migrate to urban areas in search of economic opportunities, investing in rural institutions has been identified as one of the ‘prime movers’ of agricultural development. As the agricultural sector remains one of the key employers and main sources of income for most rural households in South Africa, key areas of investment need to be identified that will have a positive impact on agriculture. There is an opportunity to analyse poverty’s relationship with other agricultural factors within the NIDS data set. We find that agriculture directly affects household consumption and hence overall poverty and welfare.