Investigating the relationship between computer literacy and job vulnerability during the coronavirus pandemic in 2020: evidence from South Africa

Abstract

Despite the acceleration of digital adoption in the workplace with the coronavirus pandemic, there have been few studies on how and which workers have adapted to this in South Africa. This study investigates the relationship between computer literacy levels, ICT ownership, and the probability of employment during the coronavirus pandemic in 2020. Using data from Wave 5 of the National Income Dynamics Study and Wave 1 and Wave 2 of the National Income Dynamics Study – Coronavirus Rapid Mobile Survey, this paper compares computer literacy against employment outcomes during the month of April 2020, when South Africa was in a stringent lockdown, and June 2020, when the economy began to open up. This paper restricts its estimations to those employed in February 2020, to observe whether computer literacy and ICT hardware ownership protected their employment during April 2020 and June 2020. The results show that there is considerable variation in computer literacy levels increasing the likelihood of employment in April 2020 and June 2020. Owning a computer and having a post-matric education increased probabilities in April, while computer literacy was not shown to protect employment much during the hard lockdown. As the economy opened up in June and firms began to digitally adopt, computer literacy played a positive, significant role in increasing the likelihood of employment. However, when observing by subgroup analysis, we see that those with a post-matric education did not actually benefit as much as expected from being highly computer literate. Furthermore, while female employment probabilities were generally negative overall, females with a matric benefit most from computer literacy in June, which is in line with Casale and Shepherd (2020) who showed that women saw slightly larger employment improvements compared to men from April to June. In addition, Ranchhod and Daniels (2020b) observed that from April to June, there was an increase in the percentage of individuals with a matric employed, and suggested that less-skilled workers who were employed in February and lost their jobs in April were able to recover in June. This suggests that, in addition to sectors reopening, the adoption of task-based work shifted in this period to being conducted remotely and digitally, and digital platforms being associated with employment creation, which may require less-skilled workers to have some computer literacy skill.