Is it possible for Eskom to fail?

When I asked myself, will Eskom fail? My first thought is that it is failing already, but of course the real issue is: can it be turned around? And, what of the budget speech? Will the new incentives make a difference?

So, can Eskom be saved?

In theory, yes — with a government determined to enforce its own laws and the right people in charge, it could be done. After all, Ukraine operates power infrastructure under direct attack in a war. So anything is possible. But the odds are stacked against an Eskom turnaround.

Why?

Infrastructure looting is the crack cocaine of corruption. Nothing is more addictive or more self-destructive.

The more essential the infrastructure, the worse it is. Why? Because the response to failure is to throw more money at the problem. And that creates a perverse incentive to fail if looting is your game. The worse the problem, the higher the cost of repair and the greater the looting opportunity. Contrast this with other less essential state-owned entities. SAA is a shadow of its former self. 

Despite the failure of Comair — which operated the domestic brand of BA as well as its own Kululu — there are other options. So when SAA became increasingly unviable, it could have potentially been shut down. The Post Office is going the same way. It never really served the poor, who often have no street address. The better-off can use couriers for packages and the internet for letters and documents. So the failure of the Post Office, while sad, is not crippling.

But when it comes to daily essentials — electricity, water, sanitation and transport — things cannot be left to just fall apart. Yet, this is what is happening. 

Even formerly well-run cities are increasingly falling apart, with widening potholes and water outages. I have written about infrastructure challenges in Makhanda, like hundreds of millions of rand spent on water plant upgrades that are never completed. Many other cities will have similar issues. The worse things get, the more money is thrown at the problem — yet it gets worse and worse.

If the sole aim is to feed corruption, it’s the gift that keeps on giving. Since the infrastructure is essential, its absence feeds an investment strike. Who is going to invest in productive capacity if infrastructure is falling apart? So the desperation to fix the problem grows, you get emergency measures to speed spending with even fewer checks and balances than normal. It’s a feedback loop: more failure means more money, which means more to loot.

Let’s get back to Eskom. Back when the proposed nuclear build surfaced in 2016, R1 trillion seemed like an enormous, unaffordable sum. There are other reasons to be suspicious of this deal — the way it was negotiated behind closed doors and the hidden costs of nuclear (waste disposal, dirty mining, the link to weapons). 

However when you line up the cost against what Eskom has been paying for its two newest coal plants Medupi and Kusile, this is not such new terrain. One estimate in 2019 put the total cost at that point at R694 billion and that wasn’t the end; in 2022 alone, a further R33 billion was added. Madupe was started in 2007 and Kusile in 2008, each with an initial budget of about R80 billion. 

Massive cost overruns are commonplace with nuclear plants but not as much with coal; a final cost of around five times the initial budget is pretty extraordinary. And that assumes there will be a final cost.

Madupi and Kasule together add up to about 9 500MW, close to the amount proposed for the nuclear build.

So would nuclear, despite its risks, be a solution?

As long as the government does not give up its addiction to looting infrastructure projects, a nuclear build is bound to run into massive overruns both in time and cost. Even if that is somehow avoided, these are big, complex projects that can take ten years or more to complete.

So nuclear will not solve Eskom’s immediate crisis.

What of the solar and renewable incentives in the budget?

On their own, they could add a lot of capacity to the grid fast. I will be fascinated to see if the Treasury has accurately estimated the cost to the fiscus of the tax concessions.

Private households score up to 25% of the cost of solar panels. The government budgets R4 billion for household solar panels, indicating at least R16 billion will be installed, based on the 25% rebate, up to a maximum of R15 000. As a rough estimate, based on pricing a range of solar panels, R7 million buys about 1000MW, so the budgeted subsidy equates to nearly 600MW of panels.

For business, renewable power can be claimed as a 125% deduction against income in one tax year.

The business tax incentive scheme is not specific to solar but other renewable energy sources like wind turbines are more complex to build and not a likely choice for someone wanting a project up and running by the end of the tax year, the target for the tax incentive scheme.

The cost is budgeted at R5 billion. Since the corporate tax rate is 27%, the saving to a business equates to 33.75% of the cost of the whole system, so a cost of R5 billion to the government adds up to nearly R15 billion of installed cost. The fraction of a system that is solar varies a lot. How much battery is there, for example? Let’s take 20% as a ballpark. This means the budgeted amount equates to about R3 billion in solar panels or about 400MW.

Add up both incentive schemes and you get about 1GW, adding about 2% to South Africa’s nominal generating capacity (53.7GW in 2021).

Each stage of load-shedding aims to reduce consumption by 1MW so an additional gigawatt on the grid will make a huge difference — though exactly how huge depends on how much storage goes with the new capacity, since solar varies a lot during the day and is off at night. CSIR in 2021 estimated solar countrywide as supplying 26% of its rated capacity on average (the capacity factor). 

That figure may be on the high side if solar becomes more widely installed including in areas of lower average sunshine. Even so, this scale of addition to the grid will be significant and can happen fast in thousands of small projects, very different from one or two megaprojects that take years or even decades to deliver with a high chance of failure.

The biggest risk to the government of these schemes is that they may be in much more demand than budgeted. However, if a huge increase in solar power reduces downtime, the gain in increased productivity — and hence tax revenues — could offset any increase in the budgeted cost to the government.

Longer term, a big increase in wind power would be helpful as it is less dependent on storage than is solar.

All of this however is only part of the solution. The beneficiaries of these schemes are the relatively wealthy. A very small business making little profit cannot benefit from a generous tax deduction. Private households of limited means will not benefit from a 15% discount on solar panels, only part of the system cost. A programme is also needed to put solar power on schools, hospitals and other public facilities that mainly service the poor. This will reduce running costs for these facilities, increasing their resources for providing services. There should also be a programme for subsidised solar power in low-income and RDP homes.

Local governments should bid for government funding to back up their essential infrastructure including water, street lights and government buildings. This has two benefits: services are less impacted by load-shedding and reducing their own Eskom bill will compensate for less income from residents and businesses taking advantage of the new incentives.

Sadly, anything administered by the government is at risk of looting. For this reason, I would like to see trusted NGOs running programmes for the poor. However, more reliable basic services will benefit the poor so equity is not totally cut out if the local government takes up my proposal.

So, where does this leave Eskom?

We are still far from completely replacing its fleet. Even 1GW of solar is only as much as the smallest coal power plant, with a much lower capacity factor. But if Eskom carries on in its downward trajectory, the incentives to look for alternatives increase. The biggest risk is that we end up with a Model C country: those with the means pay for what should be a universal service and those without struggle. Is this what we really want?

A more rapid move to renewables should be part of achieving climate justice, not yet another way to disadvantage the poor.

Back to the original question: does any of this help Eskom? It may by taking off the immediate pressure to do something urgent. Without that pressure, more considered decisions can be made to turn Eskom around, with less pressure for emergency funding – and hence more oversight.

Ironically, the greed of those who have captured the state and Eskom in order to maximise profits from coal could end up shortening the life of coal. Some imagination is needed to turn this into an opportunity for equity, not yet another amplification of inequality.

The views expressed are those of the author and do not necessarily reflect the official policy or position of the Mail & Guardian.