Few questionable economic analyses and disproven solutions to South Africa’s economic and socioeconomic problems deserve responses. However, when the writer is an experienced business leader of Colin Coleman’s calibre, with a history of leading business at the heart of South Africa’s economy, a response backed by facts and objective realities of the country’s economic problems is warranted.
I respond to Coleman’s remarks at the National Investment Dialogue, in which I participated, and his Sunday Times column of 3 October, titled “SA doesn’t have a debt problem. It has a growth problem – and a solution”.
Coleman’s main conclusions are that 1) South Africa’s debt of R4-trillion (71% of GDP and rising) is not a problem, 2) The country has a growth problem (having averaged 1.7% from 2010 to 2019 and 1.0% from 2010 to 2020), and 3) the solution to South Africa’s growth problems is a stimulus package through an employment incentive scheme (or a basic income grant), and improving completeness, productivity and reducing the cost of business.
The fourth conclusion is implicit, and it is that we can finance the stimulus by borrowing more debt without negative consequences if economic growth rises between 2.5% and 3.0%.
Yes, South Africa has a growth problem
I partially agree with Coleman’s second conclusion that South Africa has a growth problem; however, the evidence does not support his view that the cause of low growth is lack of stimulus, if we take the US and China’s stimulus as examples.
Furthermore, his other conclusions, that South Africa’s R4-trillion debt is not a problem, and that we can borrow more to institute an employment incentive programme as a growth stimulus without having to worry about debt, is not supported by the country’s experience and data.
First, let me provide the reality of South Africa’s growth track record relative to its true peers, and not just the US and China, economies that are very different to ours, which Coleman uses to justify a state-led growth. From 2000 to 2009, South Africa’s trading partners’ trade-weighted economic growth averaged 4.5% while the simple average growth was 4.2%.
Emerging and developing markets’ economic growth averaged 6.1%. South Africa’s average growth over the same period was 3.6%. Much of this growth for South Africa and other commodity-exporting, emerging and developing economies was driven by the supper commodity cycle, whose end was triggered by the global financial crisis of 2008.
For the decade that followed, from 2010 to 2019, the trade-weighted growth of trading partners affected by the global financial crisis averaged 3.8%, just marginally higher than the simple average of 3.7%. Emerging and developing economies averaged 5.1%. By contrast, South Africa’s average growth rate more than halved to 1.7% from its average annual growth rate in the prior decade. There is therefore no doubt that South Africa has a growth problem compared to its trading partners, and emerging and developing economies. This is where the agreement with Coleman ends as far as his main conclusions are concerned.
Commodity, climate change, coal and platinum
Second, let me turn to aspects that largely explain what changed in the South African economy between 2010 and 2019 that choked the country’s growth more than its peers. The first factor is a global and structural shock to commodity prices, which have helped lift economic growth in the previous decade but began to act as a drag on economic growth since 2014.
South Africa’s exports are largely commodity raw materials that are not sought after and are weakly integrated into global and regional value chains. Thus, when commodity prices started to decline, South Africa’s economic growth decoupled from global growth.
One of the most troubling views from Coleman, when he says South Africa now has the wind at our backs globally, has to do with his commodity price outlook. He says the commodity price cycle we have seen since the recovery from the Covid-19 pandemic is not a flash in the pan, nor is it a post-pandemic bounce-back, but a secular long-term response to a changing global economy driven by climate change.
Climate change concerns mean that coal, one of South Africa’s major export earners, will experience a structural decline over the coming years. Following the diesel automotive companies’ emission scandals in 2015, many countries such as Denmark, the Netherlands, the UK and Ireland plan to end the sale of fossil-fuel-powered cars by 2030. California plans to do so by 2035 and France by 2040.
This trend is clear, more countries are likely to follow, and it will mean that platinum, a major export revenue earner for South Africa used in the auto catalytic converters for diesel cars, will not be one of the commodities that will form the super cycle that Coleman alludes to.
There will likely be more global demand for commodities that are linked to greening the economy, however, that we are on the verge of a multi-year commodity super cycle is not assured. The World Bank commodity price forecasts show that on average between 2022 and 2025, coal, zinc, lead, tin, iron ore, silver and gold will all decline each year. Aluminium and platinum are projected to rise by an average of 1.5% and 1.0% per year — this can hardly be a commodity super cycle.
To base a commitment to permanent spending like an employment incentive scheme (or a basic income grant) on a source of economic growth and tax revenues that is highly uncertain, like a future commodity super cycle, will be a major policy mistake.
Debt is a problem
The tragedy of this view is that politicians facing local government elections in less than a month and a national general election within two years, with a misfiring economy and record high unemployment, will find that idea appealing.
Unsuspecting politicians are wrongly led to believe that South Africa’s economic growth and, therefore, job creation, is driven by lack of consumer demand, and that if we provide an R800 employment incentive or wage subsidy to the 12-million unemployed people, companies will hire more people and the economy will fire on all cylinders.
Politicians are also wrongly led to believe that debt is not a problem, thus we can borrow to fund this R115-billion stimulus without a debt problem in the hope that growth will pick up to between 2.5%-3.0%, arguably driven by household consumption of food, clothing and essentials.
Sure, the retail giants would love that. They have a stable demand from the child and old-age support grants and will be over the moon to get another source of demand from the stimulus. The shareholders of these companies will also celebrate as the value of their shares will go up. But that is just about it. The goods sold by these retailers are mostly imported, particularly manufactured goods, thus a significant share of jobs created with this stimulus will be exported.
Economic growth might not actually rise as projected and the leakage from the imports will mean tax revenue collections will not rise enough to self-finance the increase in debt. In line with this, the other implication of the revised national accounts is that tax buoyancy, the amount of tax generated for each rand growth in the economy, is lower than previously thought, which further deems the view that the stimulus can be self-financing in the long run.
As far as a wage subsidy is concerned, there is recent evidence from the wage subsidy that was implemented by the government in 2014. Professor Haroon Bhorat and others in their 2020 study of the employment tax incentive scheme make several findings that contradict a wage subsidy as a solution. They conclude that larger firms take up the employment incentive more than smaller firms. The employment incentive supports a larger number of jobs, but this is not equivalent to creating jobs. The evidence remains mixed where jobs have been created.
At the aggregate level, they find that the wage subsidy has no impact or has a positive-but-small impact on employment levels. It follows then that a sustainable solution for job creation in a fiscally constrained economy cannot rest on a wage subsidy; a subsidy will simply be a cash transfer to participating firms.
The third and more specific factor, which is the subject of the current national debate, is social spending. Let me begin by narrating a simplified life of a poor household. A child of non-schoolgoing age receives a child support grant from the government, thereafter the child goes to basic schooling financed by the government, followed by post-matric schooling now financed by the government. Healthcare, safety and security are all provided for by the government. Part of the transport expenditure is financed by the government through subsidies to the taxi industry.
The total social spending viewed from this functional spending lens amounted to R687.2-billion in fiscal year 2010-11, which is 78% of the total R880-billion consolidated expenditure. In this I included social protection, education, recreation and culture, health, housing and community amenities, environmental protection, transport, public order and safety and defence.
By fiscal year 2019-20, social spending had doubled to R1.4-trillion, accounting for 76% of total consolidated spending, which is a decline as a share of total spending.
Over the decade, the government protected the most important social spending (social protection, health and education) in the consolidated budget. As a share of total spending, social protection increased from 14.8% to 16.1%; education rose from 19.4% to 20.5%; and health rose from 11.7% to 12.1%.
To enable this protection of core social spending increase in an environment of low growth and rising debt service costs, which displaced social spending and to which I return to next, reductions in the share of total spending were made in housing and community amenities from 11% to 9.4%; transport (6.7% to 5.0%); public order and safety (from 9.6% to 8.5%); and defence (3.5% to 2.7%). This is hardly a budget that is anti-poor nor is it a budget that treats the marginalised and unemployed with cruelty.
Fourth, where is debt in all these reductions in social spending? Simply put, the reason the government had to reduce the share of spending on housing, transport, public order and safety and defence is because South Africa’s gross debt rapidly rose by more than three times from R990.6-billion in 2010-11 to R3.3-trillion in 2019-20.
This resulted in South Africa’s credit ratings being downgraded to sub-investment, and consequently debt-servicing costs significantly rose from R66.2-billion in fiscal year 2010-11 to R204.8-billion by fiscal year 2019-20. The share of debt service costs in the total consolidated spending went from 7.5% to 11.2% over the same period and is projected to rise to 16.2% by 2023-24.
There is no clearer evidence to dispute Coleman’s main conclusion that South Africa does not have a debt problem than the objective reality that the country is now sub-investment-grade-rated due to a rapid increase in debt and the hard facts that social spending must be reduced to pay for rising interest on debt. The correct characterisation is that South Africa’s growth problem has a debt problem, which simply means that we have a debt problem.
Fifth, if we look at public investment spending by the government, it is also clear that the rising debt service costs crowd out investment and effectively put the government on an investment strike. As I wrote in Business Day on 1 October 2021, investment by the general government contracted by an average of 1.4% per year between 2010 to 2020.
Investment by public corporations contracted by an annual average of 3.3%. Even with the Eskom power plants-build stripped out, investment by public corporations still show negative growth. Investment by the private sector investment contracted by 1.2% over the same period. Consequently, total investment in the country contracted by an annual average of 1.7%.
If we exclude the 2020 figures, which are distorted by the Covid-19 shock, general government investment contracted by an annual average of 0.7%, public corporations contracted by 2.4% on average, while the private sector grew by 0.1%. These numbers are based on the rebased and 2015 benchmarked constant figures.
What these figures show is that the South African government has been on an investment strike over the past decade, necessitated by the rising debt-service cost due to the rising debt and the credit rating downgrades, and the need to protect crucial social spending on social protection, health, and education.
Finally, the correct economically plausible conclusion is that South Africa sacrificed investing in critical infrastructure that would boost growth through improved competitiveness, productivity and easing doing business, due in part to rising debt. By doing so, South Africa choked economic growth and job creation, thus increasing the number of unemployed people that need short-term support. A stimulus funded by debt will recreate and accelerate the crowding out of social spending and investment as debt service costs will continue to rise as they are currently projected.
In fact, looked at from a multiyear perspective, a debt-funded stimulus will blow out public finances and hand the country to the International Monetary Fund. It is the job of the bureaucrat whose term is not limited by the political cycle to look beyond the election cycle and craft solutions that will make the country’s public finances sustainable 10, 20, and 30 years from now.
Future generations must not be made to pay for the exuberant consumption of the current generation. That is the principle of intergenerational equity. A consumption stimulus financed by debt violates this and many other economic principles. All said, the slogan must be social welfare: savings and investment not debt, or taxation and consumption.