SOE debt and the implications for govt finances

Anton van Dalsen says the issue must be analysed in light of lower than expected economic growth

Guarantee obligations for the debt of state-owned enterprises, combined with limited economic growth: what threat does this pose for Government finances?


The Development Bank of South Africa (DBSA) reportedly threatened to call in its loan of R15 billion from Eskom at the end of July 2017, as a result of the latter’s lack of action on receiving a qualified audit. This sent the Ministers of Finance and of Public Enterprises scrambling to ensure that Eskom’s chief financial officer was placed on special leave with immediate effect, pending an investigation.

The standard “cross-default” loan clause would mean that all of Eskom’s debt would become repayable immediately if one lender calls default. A public spotlight has been put on an important issue: the state’s obligations in guaranteeing large slices of debt of state-owned enterprises (SOEs). The dangers have been pointed out some time now, but the DBSA action has made it all seem very concrete, very suddenly.

However, the level of guaranteed debt should not be seen in isolation. It should be seen in combination with another serious threat to Government’s finances: the general effect of lower than expected economic growth. Taken together, these two issues pose a serious threat to government’s finances. This brief outliness the risks which they present.

The IMF view

In its recent report on South Africa, [1] the IMF states that “South Africa’s government debt-to-GDP ratio is sustainable in the baseline scenario, but is highly sensitive to shocks. Persistent low growth or a shock to contingent liabilities could push both South Africa’s gross financing needs and debt level above the high-risk benchmarks of 15 and 70 percent of GDP, respectively, indicating an increased risk of financing difficulties and debt distress in those scenarios.” The percentage levels of 15 and 70 are not randomly chosen, but represent levels which experience has shown to be definite danger thresholds for government finances in emerging market economies.

In a lower growth scenario, Government revenue does not increase substantially in real terms without a drastic increase in taxation. This makes it impossible for Government to fund any unexpected obligations from its annual revenue - for example, by having to settle guarantees issued on behalf of SOEs. Expenditure on the latter items will therefore have to be funded by ever more debt, whose debt service costs mean that the available revenue for spending on normal budget items is in turn squeezed - thus placing ever greater pressure on the Government budget. This is what an increasing debt-to-GDP ratio in effect tells us.

South Africa’s current gross debt service costs are at 11.5 percent of aggregate Government expenditure and Government debt is equivalent to 52 percent of GDP, which are not seen as problematic levels.

However, the IMF report further states that “The results suggest limited room for policy response to further adverse shocks and underscore the urgent need for reforms to increase economic growth and reduce contingent liabilities, especially from state-owned enterprises, to increase the resilience to shocks. Insufficient reform progress risks a continuation of the ongoing vicious cycle of weak growth, a growing debt ratio, and the repeated need to take fiscal measures to keep the debt sustainable.”

Two of the GDP growth scenarios which the IMF applies in its stress tests on South Africa’s economy, are set out in this table (in percent per annum):

Real GDP growth scenario














Growth shock







In the “baseline” scenario, the IMF projects the debt-to-GDP to increase to 57 percent by 2022. However, in the “growth shock” scenario, gross public debt would reach 60 percent of GDP in 2019 and would then gradually increase to just under 62 percent in 2022. The IMF further finds that a persistently low growth scenario (with growth at 1 percent lower than the baseline throughout the period), would result in the debt-to-GDP ratio reaching 71 percent of GDP in 2022. This finding is premised on tax revenues not covering programmed spending (ie. excluding interest on debt) - but this would in any event be likely in a low growth scenario.

In comparison, the OECD’s economic survey of South Africa, published in July 2017, [2] forecasts that in a growth scenario of 1 percent per annum, the debt-to-GDP ratio will increase on a constant basis from current levels to 70 percent in 2030.

Guarantees by Government for SOE debt

How does SOE debt which is guaranteed by Government, fit into this discussion? First of all, let us look at the absolute numbers. Government’s guarantee exposure is the following [3]:

in R billion, as at end of

February 2017


Guarantees available


Actual guarantee exposure







Trans-Caledon Tunnel Authority



South African Airways



Land Bank












Independent Power Producers






We have previously drawn attention to the precarious financial position in which Eskom finds itself [4]. The recent DBSA action has now also focused attention on the fact that Government has committed itself to providing guarantees amounting to R350 billion on behalf of Eskom, if required. In addition, Government has confirmed that guarantees of R200 billion are available for Independent Power Producers (IPPs). In this case, the risk of the guarantees being called really amounts to the risk of Eskom defaulting on its obligations to buy the generated electricity. The aggregate potential Eskom risk in relation to Government’s guarantees therefore actually amounts to R550 billion, equivalent to approximately 12 percent of South Africa’s GDP for 2017.

As we have pointed out before, [5] Eskom is currently facing an approximate R32 billion annual cash flow deficit over each of the next 5 years and has approached the National Electricity Regulator (NERSA) for a 19.9 percent tariff increase in order to plug the forecast hole in its cash flows. [6] We do not see NERSA approving such a high tariff increase, which means that Government will have to provide additional funding to Eskom (and this will have to be funded by increasing Government’s own debt).

This would be in addition to settling any amounts that are already guaranteed and presents yet another funding challenge for Government. We are not even going to discuss a potential nuclear new build programme, for quite apart from the question as to whether it is needed or not, how could Eskom possibly fund another R1 000 billion for this venture?

The combined effect of large guarantee settlements and lower economic growth

We see it as extremely unlikely that Government would be called on to settle the full or even a greater part of the amount of guaranteed debt. However, if we keep in mind that R 50 billion represents roughly 1 percent of GDP, it becomes clear that any substantial guarantee payments by Government, would get us to the debt-to-GDP danger zone which is mentioned in the IMF and OECD surveys, if there is no meaningful GDP-growth.

As far as the growth forecasts are concerned, we see the baseline scenario assumed by the IMF as optimistic and the danger is therefore that the debt-to-GDP ratio will grow more quickly than it has predicted. Recently published short-term expectations by other institutions point to lower growth numbers: for instance, the OECD and the World Bank both recently forecast GDP growth for 2017 at 0.6 percent (ie. 0.4 percent lower than the IMF). Some local forecasts [7] take the same view as the OECD and World Bank, whereas the South African Reserve Bank’s Monetary Policy Committee has even lowered it to 0.5 percent. [8]

Our analysis therefore indicates that the possibility of large guarantee payments by Government to keep its SOEs solvent, combined with lower economic growth, is not fanciful. This combination makes for a dangerous cocktail.


We can assume that the next five months leading up to the ANC National Conference will continue to be filled with internal ANC conflict. In this situation, Government may well not be able to do what is necessary to give market confidence a boost. In addition, we are faced with extremely ill-considered actions such as those of the Minister of Mines, which have scared anyone involved in the industry.

When it comes to tackling governance and corruption problems in SOEs, Government shows little enthusiasm, unless a financial hand grenade is lobbed at it in the manner of the DBSA. For its part, the police and prosecuting authorities seem to go AWOL when corruption in SOEs needs to be investigated. Financial governance in SOEs is therefore not going to improve quickly. This just adds to the risk of Government having to bail them out at some or other stage.

This is the current state of affairs. As a consequence, we do not see business and investor confidence improving in the short-term. We therefore run the risk of being confronted with weaker growth than forecast - and as we have shown, this is not an ideal environment for Government having to bail out large SOEs.

Anton van Dalsen is Legal Counsellor at the Helen Suzman Foundation.

This article first appeared as an HSF Brief.


[1] International Monetary Fund. See also Charles Simkins’ brief on this report:

[2] OECD

[3] Treasury

[4] HSF 

[5] HSF

[6] Ibid.

[7] See, for example, Investec Bank’s June 2017 forecasts:

[8] Reserve Bank