The divergent pathways of the pandemic within South African cities

ABSTRACT

The coronavirus pandemic has had devastating effects on urban lives and livelihoods throughout the world. A major concern in the global North has been the hollowing out of central cities caused by remote working. The consequences for cities in the global South extend further and deeper because their economies are weaker, social and spatial inequalities larger, and healthcare systems more fragile. The paper explores the uneven trajectory of COVID-19 for people and places in South African cities, drawing on unique individual panel data. It shows how communities that were already the most vulnerable have been hit hardest by the pandemic, triggering hardship, hunger and social unrest. Local institutions will have to play a stronger role if society is to manage pandemics better in the future.

KEYWORDS:

• COVID-19 pandemic
• urban inequalities
• informality
• cities
• local government

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 UNCTAD (2021) estimates that developing countries will be $12 trillion poorer by 2025 because of the pandemic. A failure to roll out vaccines could reduce incomes across the South by another $1.5 trillion.

2 The NIDS was a national household survey conducted between 2008 and 2017 with a nationally representative sample of over 28,000 individuals and 7,300 households. The NIDS-CRAM was a subsample of the NIDS 2017 population. It was weighted and adjusted for non-response to represent the national population of the NIDS sample in 2017. Hence, the NIDS-CRAM reflects ‘outcomes for a broadly representative sample of those 15 years and older in 2017 who were followed up 3 years later’ (Kerr et al., 2020, p. 2). The NIDS-CRAM was conducted as a short 20-minute telephone survey, and each participant received a R20 airtime voucher per wave for participating. The NIDS-CRAM data is analysed mainly as a series of cross-sections which can be interpreted as broadly representative of South African adults. In a few instances we take advantage of the data as a panel to explore how individuals moved into and out of certain socio-economic conditions (such as employment and hunger). This offers a dynamic view of the conditions facing different sections of the population which is masked by a static evaluation of cross-sections. The balanced panel includes 3,456 individuals who were successfully interviewed in each wave, with panel weights applied to mitigate against attrition. There are inevitable differences between estimates produced using the NIDS-CRAM as a balanced panel and as a series of cross-sections.

3 It was possible for respondents to alter their perceived neighbourhood type from one wave to the next. Such switches affected only a small share of respondents and were not large enough to alter the distribution of neighbourhood types across waves. It was most apparent for individuals who classified their neighbourhood type as ‘formal residential’, which is a relatively open-ended category. It should also be noted that W1 of the NIDS-CRAM survey did not collect information on neighbourhood type. Therefore, we impute missing information on neighbourhood type for W1 from W2 which could introduce some minor discrepancies. 82% of the W2 sample indicated they were living in the same dwelling as when interviewed in W1.

Additional information

Funding

This work was supported by NIDS-CRAM (www.cramsurvey.org).