Forecasting household car ownership in South Africa : alternative models and future trends

Abstract

The paper investigates the use of an alternative household car ownership modelling approach for South African urban areas, particularly the metropolitan areas, that moves away from existing race-based classifications, but instead uses household income and spatial attributes of an area captured in terms of dwelling unit types. The model has been successfully calibrated for the City of Johannesburg, and tested for other Gauteng areas. The paper also provides limited benchmarking of South African household car ownership against published literature, in which it is illustrated and concluded that localised research on behavioural market responses is critical. Topical issues such as development density and lifestyle choices within the context of the emerging nature of household car ownership in South Africa are also investigated, although the transitional nature of the South African economy presents some analytical challenges. Finally, the paper illustrates the model's application in the urban development planning context. The model results, supported by other qualitative considerations, point to a potentially explosive growth in car ownership to be expected in historically disadvantaged areas of South African cities as middle-class incomes grow. Thematic areas for further research in the field are also identified.