Adventures in city data: An ethnographic story

South Africa is urbanising rapidly, and its economic landscape is continuously changing as a consequence. In this context, city governments and urban scientists have long called for better access to city economic data. The National Treasury has reinforced this demand, insisting that intra-city economic data is critical in order to improve planning, performance and investment in South Africa’s cities.

A wealth of data is collected by the South African Revenue Service (SARS) in the course of its routine operations assessing the tax obligations of companies and individual taxpayers. In addition to its bureaucratic purpose, this data represents an enormous potential resource for a detailed understanding of the urban economy. Until recently, this resource has been underutilised because it was not available in an anonymised and geocoded form. At a practical level, however, the significant amount of energy and time required to access, clean and align administrative datasets to make them usable is not generally understood.

This GCRO Occasional Paper presents an ethnographic account of a decade-long journey in city economic data collation by the author who, as a long-term technical advisor to the National Treasury’s Government Technical Advisory Centre (GTAC), led the work on the city economic data programme in support of the first phase of the National Treasury’s Cities Support Programme (CSP).

After observing the critical need for anonymised and geocoded economic administrative data in policy formulation and urban research, this paper examines the reasons for the limited availability of datasets able to show the location of economic activity and employment at a disaggregated local level. The paper details how the National Treasury’s collaboration with the World Bank in 2016 to produce the Urbanisation Review of South Africa stimulated and directed the efforts of GTAC and the Economies of Regions Learning Network (ERLN) to pursue official sources of city-level administrative data. The paper goes on to recount subsequent National Treasury/CSP collaborations with Statistics South Africa, SARS and the Unemployment Insurance Fund (UIF) to collect and collate anonymised and geocoded city economic data from sources other than national general surveys. Despite progress, these efforts were ultimately stymied due to practical and governance constraints. Nevertheless, in a parallel process, these collaborations ultimately bore fruit in the establishment of a secure administrative data centre at the National Treasury that stores anonymised data, which can then be geocoded using postal codes. This secure data centre in turn, after the author had left the process, ultimately provided a foundation for the milestone publication of the 2021 City Spatial Economic Data Reports.

The paper concludes by reflecting on the insights from this ethnographic account around possibilities for improving the integrity of the city spatial economic data resource, and enhancing its use in credible, evidence-based urban analysis. First, these conclusions highlight broader institutional and public management concerns in the current governance environment on which future steps to improve the city spatial economic data will depend. Second, the paper points out that, despite the long journey travelled, business classification uncertainty still remains. Solving these governance and data puzzles may further enhance the incredible potential that such a rich data resource holds for evidence-based policy aimed at creating a more just and equal society in South Africa.

Recommended citation: Robinson, S (2022). Adventures in City Data: An Ethnographic Story. GCRO Occasional Paper #20. Gauteng City-Region Observatory, December 2022. https://doi.org/10.36634/RPLR3362