New research indicates that the largest component of South Africa’s Presidential Employment Stimulus (PES) programme is beneficial to economic growth as it also has an indirect positive effect on non-programme employment and wages.
The study, commissioned by the Presidency, investigated the domestic factor income effects of the Presidential Youth Employment Initiative – Basic Education Employment Initiative (PYEI-BEEI).
The Covid-19 pandemic has exacerbated already high levels of unemployment, deepening poverty and heightening levels of hunger and food insecurity in South Africa. Youth unemployment is particularly alarming, with 45% — or 4,7-million — youths aged 15 to 34 estimated as unemployed in 2023, according to Statistics South Africa.
To mitigate the combined health, social, and economic crises stemming from the Covid pandemic, President Cyril Ramaphosa announced, in April 2020, a range of support measures to mitigate their impact, including emergency social protection and employment stimulus measures.
Established in October 2020, the Presidential Employment Stimulus (PES) aims to create jobs and strengthen livelihoods, supporting meaningful work while the labour market recovers and augmenting existing commitments to these outcomes. The PES provides funding for several Presidential Youth Employment Intervention (PYEI) components, and directly delivers employment opportunities for the youth.
The Basic Education Employment Initiative (BEEI) is part of the Presidential Employment Stimulus programme. It is run through the Department of Basic Education and targets 18- to 35-year-olds. High school graduates are eligible to apply as Education Assistants (about two-thirds of participants) to help teachers in classrooms, and those without a high school qualification are eligible as General School Assistants (about one-third of
participants) to help with miscellaneous school tasks such as infrastructure maintenance.
The research paper, entitled “Stimulus effects of a large public employment programme”, focuses on the domestic factor income effects of the PYEI-BEEI payments, characterising the “first stage” of a possible multiplier mechanism of the PYEI-BEEI programme.
It was funded by Agence Française de Développement (AFD) and the European Commission through the EU- AFD Research Facility on Inequalities, and was undertaken by the University of Cape Town (Southern Africa Labour and Development Research Unit – SALDRU).
Researchers characterised the spending and factor income effects of the PYEI-BEEI programme. In partnership with the privacy-preserving data collaboration firm Omnisient, they matched securely anonymised participant IDs with anonymised weekly individual-level sales data from one of the country’s largest retailers and estimated the treatment effect on participant spending at the retailer.
The researchers found a sharp increase in spending of 15,4% explained by increasing consumption expenditure. They also found that this direct effect on consumption and spending was sustained for the duration of the PYEI-BEEI, and then decreased to about 4% in the months after the programme.
The researchers extrapolated the effects of the increase in retailer sales on domestic factor incomes and suggested that this phenomenon implied benefits to value added, job creation and wages in the local areas in addition to the actual payments of the PYEI-BEEI programme.
Indeed, their estimates show that the PYEI-BEEI programme would create around 1 800 additional jobs per month and would increase local community incomes by around R13,3-million in addition to the actual PYEI-BEEI payments.
The researchers conclude that domestic factor income effects may be important in evaluating the economic efficiency of government spending programmes, and may help reconcile some of the perceived trade-off in public objectives between poverty reduction and private sector jobs.
Finally, they suggest that the sales to domestic factor income conversion estimated in this paper for the PYEI-BEEI programme implies that other public spending (social grants for example) may have similar initial stimulus effects.