Menu Close

South Africa’s public service: real spending is falling, but demand is growing

Winnie Ngwekasi Primary School in Soweto, South Africa. Public schools have been under pressure since this picture was taken in 2009. Stephane de Sakutin/AFP via Getty Images

Core public services – basic education, healthcare and criminal justice – account for a large share of the consumption basket of poor South Africans. They are provided largely free of charge to any user. The public provision of these services is widely recognised as a cornerstone of social and economic development in any society.

Together, they account for three-quarters of government’s wage bill and half of spending on goods and services – the medicine, books and cars that doctors, nurses, teachers and police officers need to do the job. Over the last decade, there has been a chronic and deepening erosion of the resource base on which public services depend. Users of these public services have been caught in the middle of government’s need to consolidate the fiscus on the one hand, and the demands of public sector unions for better pay on the other.

In a recent report, we appraise the choices made in the government’s budget statements, and numbers that are tabled as part of those statements. We argue that the fiscal consolidation as currently proposed will significantly reduce real spending on core public services. It will erode the quality and reach of these services, and widen income inequality in South Africa. These choices are at odds with the constitution. They will certainly lead to a retrogression in socioeconomic rights.

Government has provided no evidence that suggests a contrary conclusion. Nor has it presented any plans or policy interventions to alleviate the damaging effects of fiscal consolidation on public services.

The report gauges the quantitative aspects of public policy in three respects. First, we identify the trends in real spending over the last two decades. Second, we use the budgets approved by parliament and provincial legislatures to gauge the impact of budget choices on real resource allocation over the next three years. Third, we present analysis of government pay and employment trends, which are strongly concentrated in the core public services that are our interest.

Real spending is falling

We find that over the last decade, there have been significant reductions in the real value of basic education and criminal justice. Healthcare budgets have been under increasing pressure.

In basic education, government spent about R20,000 (about US$1086) per learner in 2009, but this had fallen to about R16,500 per learner by 2021. If provincial governments’ budgets are executed without adjustment, the next three years will see a large negative shock to the real value of spending per learner. In a worst-case scenario spending could fall to R14,000 per learner.

Government currently employs one educator for every 33 learners enrolled in the public school system. This could rise to as high as 39 over the next three years because the budget can only be realised with significant reductions to employment in the sector.


Read more: A teacher retirement wave is about to hit South Africa: what it means for class size


In healthcare, expenditure has stagnated in real terms relative to the population who depend on government services. In 2012, there were more than 720 healthcare sector workers per 100,000 uninsured people. This ratio has steadily fallen since then, reaching 632 by 2018. Spending and employment increased in response to the COVID-19 pandemic in 2020. But current budgets imply a reversal of these increases, and a reduction of spending per capita being to a historical low. Healthcare workers per 100,000 citizens could fall as low as 590.

Given the systemic inequalities within the healthcare systems, this shock is likely to be unevenly distributed. Provincial, specialised and district hospitals are likely to face the brunt.


Read more: South Africa's waning economic fortunes demand harsh policy adjustments: who will bear the brunt?


For the criminal justice sector, our analysis shows that by 2010 the level of spending had increased to more than R2,000 per citizen. By the time the COVID-19 pandemic hit in 2020, however, spending had fallen below R1,700 per citizen.

Police employment reached around 200,000 in 2010 but was reduced by around 15,000 personnel by 2020. If current budget plans are executed, police spending and employment levels will fall even further in the years ahead, reaching their lowest point over the last 20 years.

Cutting employment means cutting services

The 2022 budgets of national and provincial governments imply large reductions in government employment in all these services under any reasonable assumptions about pay improvements. Even if the assumptions made in the budget for average pay increases of 1.5% a year were to be realised, large and damaging headcount reductions are required to meet expenditure limits.

We show that the government wage bill is dominated by core public services. The professionals who provide these services – teachers, doctors and nurses, and police officers – dominate government employment. Within the core public services, the balance between professional and administrative staff appears stable and sensible.

“Bloating”, if it exists, is concentrated in political and executive offices, economic regulation, infrastructure services and public administration – particularly finance and co-operative government. These have seen substantial increases in employment in recent years. Even so, total employment in all public administration departments was less than 40,000 in 2019 compared with more than 1 million in education, healthcare and criminal justice.

From our bird’s-eye view of the government’s payroll data, there is little evidence that the employment structure is deficient – a widely held view in public discourse. The implication of this is that fiscal consolidation will lead to a further withdrawal of core services, rather than an improvement in efficiency.

Even if government could find efficiencies, reduce unnecessarily “bloated” bureaucracies, or overcome wasteful spending and corruption (and it has tabled no clear plans to do this), the currently planned path of fiscal consolidation would still largely depend on reducing the real value of core public services.

The emphasis of government’s programme is to reduce average pay. It is sometimes believed that government employees are overpaid and unproductive, and therefore reductions in their numbers and pay can be achieved without negative impacts on public services. Evidence presented in our report questions these assumptions.

It is true that over the last 20 years, most government employees have enjoyed significant improvements in pay. However, these improvements are strongly concentrated in the period 2007-2010. Since then the average pay of most government employees has grown at a moderate pace, largely in line with pay trends for similar workers in the private sector. Forcing real incomes of government workers below their private sector counterparts could erode the human resource base of public services.

Compensation spending cannibalising budgets

For many years, budget allocations have not kept pace with pay increases agreed to by government. Spending on compensation of employees has been contained within strict limits for many years. In effect, Cabinet has been increasing civil service pay while adopting budgets that effectively invalidate its own decisions.

These pressures have resulted in three forms of “crowding out” that have eroded state capabilities:

  • Most damaging has been the reduction in the number of employees. Employment in core public services has fallen relative to broad measures of public demand for services (such as the size of the population or enrolled users). In the case of basic education and criminal justice, there have been absolute falls in employee numbers.

  • The real incomes of government employees who fall outside the bargaining unit – senior managers and judges – have been forced down consistently over the last decade. This, combined with successive bargained agreements that give higher pay increments to lower-level employees, has contributed to a compression in the government wage structure. In a context of rapid increases in pay for private-sector executives and others at the top end of the distribution, this is likely to have contributed to the brain drain from the public service and the operational collapse experienced in so many government departments.

  • Where government departments have been unable to keep compensation spending within the imposed limits, they have reallocated budgets away from procurement, capital and maintenance spending, towards compensation, which now accounts for a larger share of budgets. This has been particularly marked in healthcare and criminal justice.

While spending on government consumption has been held down over the last ten years, demand for public services has increased substantially. Meanwhile, as public employment in health, education and criminal justice has stagnated, employment by private companies providing same services has surged.

This shift may be welcomed by some as contributing towards more efficient services. However, rising private sector provision of social services in tandem with deep cuts to public services means a redistribution of resources from the poor to affluent citizens. This could materially widen South Africa’s extreme levels of inequality.

The need to negotiate a new path ahead

We suggest the following change in direction:

  • First, the government should announce a programme that reflects clear policy choices and explicit plans to balance fiscal consolidation with the need to safeguard frontline services by reducing resources for lower priority programmes and departments. It should ensure that employment of teachers, nurses and doctors, and police officers keeps pace with the demand for services.

  • Second, government should negotiate a longer-term agreement with its employees that balances the need for decent pay, sufficient employment and measures that improve the effectiveness of public services and enhance productivity. This should form part of a programme to improve public provision and restructure public services.

  • Third, a fiscal rule to anchor expectations of growth of the salary bill should be considered in the context of social dialogue. This would enable a better alignment between fiscal objectives and the need to sustain the services which form a critical part of South Africa’s fragile social and political compact. A fiscal rule of this nature would need to command broad support from the public and acceptance of public sector unions.

This is an edited extract from the report, Public services, government employment and the budget, published by the Southern Centre for Inequality Studies.

Want to write?

Write an article and join a growing community of more than 181,000 academics and researchers from 4,921 institutions.

Register now