COVID-19 and the labour market: Estimating the employment effects of South Africa’s national lockdown

Like most around the world, the South African government quickly implemented a relatively stringent national lockdown in response to the COVID-19 pandemic. Although much research documenting the pandemic’s economic effects exists, these studies are largely descriptive in nature and are therefore unable to distinguish changes in employment attributable to lockdown policy versus other pandemic related factors such as foreign policy and consumption-related behavioural responses. In this paper, we seek to specifically isolate and provide causal evidence on the effect of South Africa’s lockdown policy in particular. To do so, we adopt a quasi-experimental econometric technique to exploit variation in legislated industry-level permission to work and the coincidental timing of the lockdown and data collection dates of nationally representative labour force data. We find that the national lockdown decreased the probability of employment for those not permitted to work by 8 percentage points relative to the control group. This significant, negative effect holds across several robustness tests. Using this estimate we can approximate that, of the 2.2 million fewer people employed, South Africa’s lockdown policy directly accounted for just under 600 000 (or 26% of total jobs lost), suggesting the majority of job loss can be attributed to other pandemic-related factors. We further show that the lockdown particularly jeopardized the livelihoods of those in the informal sector, with an estimated effect nearly 3 times larger than the overall effect. The vulnerability of this group to the economic consequences of the pandemic is of concern, given that their informality presents a challenge for government to provide targeted relief. To prevent further widening labour market inequalities, our analysis emphasises the importance of effective policy to support the livelihoods of those in the informal economy.