It is more than four years since Eskom asked the government to help relieve it of its unsustainable debt burden, and the government undertook to do so. Now the Eskom debt issue is on the table again. Finance minister Enoch Godongwana has made it clear the government plans to take on a portion of the Eskom debt. He will outline the principles in his medium-term budget later this month.
Markets will be closely watching the how much and the how. But the why is not altogether obvious. Eskom doesn’t generate enough cash to service its debt of almost R400bn; it has long argued that it needs to get about half of this off its balance sheet.
But instead of a “big bang” solution the government instead simply handed over almost R170bn in cash over four years to enable Eskom to pay the interest and repay capital; and in the absence of a solution would continue to transfer about R22bn a year to the ailing power utility.
So why a big bang solution now? A first reason is the cost of debt, not just for Eskom but for the government itself. The government guarantees most of Eskom’s debt and the long uncertainty over its fate has added to the risk premium it has to pay on its borrowing; addressing the Eskom overhang could therefore make it cheaper for the government to borrow.
A second relates to the unbundling of Eskom. In particular, if the new independent transmission entity is to work as it should to enable new independent renewable energy producers, it will need to invest in strengthening the national grid and in backup storage. That will require a sound balance sheet it can use to raise the funding it needs. For now, R40bn of Eskom’s debt has been allocated to the transmission entity; cutting that to say R20bn would put it in a much stronger position to borrow in the market.
But a third and perhaps most compelling reason is that, as RMB Morgan Stanley economist Andrea Masia puts it, now is an opportune moment. The government’s finances are looking far better than anyone expected even at the time of February’s budget thanks to the commodities boom and the large revenue overrun that’s resulted, as well as to the fact that the government has kept expenditure under control.
Masia argues that the windfall revenues present a unique opportunity to address the long-standing uncertainty around the Eskom debt. He calculates that if the government were to take R250bn of Eskom’s guaranteed debt onto its own balance sheet in fiscal 2023/2024 it would lift the public debt ratio to 72%, from the 68.5% it would otherwise have been. Gross debt would reach 73.2% in 2026/2027 instead of 70.5%.
This is far lower than the worst-case scenario of 80% the Treasury was projecting just a year or two ago. Masia expects the lower risk premium would help borrowing costs over time. And he points out that while an additional R250bn of debt would add about R10bn a year to the government’s debt service costs, it would save more than half of the R43bn of cash it was going to transfer to Eskom over the next two years.
Godongwana is expected to outline “broad brush” principles in the medium-term budget and may not even announce how much of the debt will be transferred. It’s all a rather delicate process given that even to say the government is transferring the debt could constitute a default and trigger covenants. There will have to be consultations with lenders and bondholders over whatever the government proposes before clearer details can be announced in the February budget.
For Eskom it’s not just a question of how much but which debt it can hope to get off its balance sheet — and how. The timeline is also one to watch — chances are the debt relief package won’t be in place until next year, and it may be implemented over two to three years.
The ideal will be to take a look at the whole debt portfolio with an eye to where debt relief is easiest to execute and what debt Eskom would like to keep — or shed. So, for example, most in the market are assuming it’s the guaranteed debt that is in line for relief. But some of that is long term, cheaper debt that Eskom would prefer to keep — for example from the foreign and local development finance institutions that account for just more than half of the R400bn, and who might in theory be easiest to negotiate with.
Conversely, some of the unguaranteed debt, especially foreign bonds, is very expensive debt the power utility would love to get rid of. One of the easiest options is simply for the Treasury to redeem Eskom bonds as they mature — as R40bn of foreign (unguaranteed) and local bonds will during 2023. Other options will be to swap particular bonds or loans for sovereign bonds or loans, by agreement with the lenders. Yet another option is to redeem some of the debt early.
It might look easier to negotiate with a big bondholder such as the Public Investment Corporation, which holds 57% of Eskom’s domestic debt, rather than with multiple investors. But any deal involving publicly listed bonds would have to be fair to all bondholders, so the methodology will be crucial.
The big picture question is whether a debt relief package for Eskom is any sort of solution to its debt problem. The answer is no, not unless several other things happen at the same time. The priority for Godongwana will be to put a package in place to ensure Eskom doesn’t come back to the Treasury with the begging bowl yet again in a few years’ time. He will surely attach stringent conditions that Eskom will have to meet.
Unless Eskom improves its operating efficiency, cuts its cost base and collects the R50bn it is owed by municipalities, it will just max out the proverbial credit card again and be back at the Treasury’s door in the not too distant future. But Eskom on its own cannot ensure a durable solution. That also needs the energy regulator to come to the party with a viable electricity tariff model, and policymakers to support an Eskom and electricity sector that can keep the lights on. This can’t wait another four years.
• Joffe is editor-at-large.
Would you like to comment on this article?
Sign up (it's quick and free) or sign in now.
Please read our Comment Policy before commenting.