Total factor productivity of urban agriculture on the urban periphery of Cape Town

Working Paper Number: 
Martin Dyer, Richard Mills, Beatrice Conradie, Jenifer Piesse
This paper investigates the efficiency relationships between inputs and outputs of urban micro-farms in two of Cape Town’s townships: Nyanga and Khayelitsha. The inputs in this study were land, labour, seeds and seedlings, compost and farmer experience. Data Envelopment Analysis (DEA) was applied to 33 producers supplying a social enterprise box scheme, thereby generating individual efficiency measures relative to best practice. The DEA results revealed an average level of overall, technical and scale efficiency of 72.4%, 79.7% and 90.6% respectively. Overall efficiency was negatively correlated with land holdings and the use of compost and seedlings. This is supported by the finding that the nine best-practice farms were characterised by a smaller scale of production, indicating that efficiency losses are experienced as greater quantities of inputs are used. In terms of area differences, Nyanga farms exhibit significantly higher technical efficiency, whereas farms in Khayelitsha are more scale efficient. Standardised input and output data show both the expenditure on compost and seed to be profitable, but we failed to show that mulching or operator experience increases profitability. Fully efficient farms are R2,600 per plot more profitable than inefficient farms while farms that need a windbreak earn R700 less per plot per season than more sheltered operations. These results are the first of their kind for South Africa and lay the foundation for more effective extension to the sector.