Op-Ed: Eskom has us over a barrel. Time for a different approach.

Although energy policy and planning belongs to the Department of Energy which is supposed to produce an Integrated Resource Plan (IRP) with regular updates, the actual modelling of the IRP effectively happens within Eskom itself. It is widely known that the never-ending delays in the next version of the IRP is due to the fact that in order to retain the integrity of the modelling process, procuring nuclear energy just won’t work. That is the wrong answer for the politicians who have to send the finding back in the hope that somehow it will come up with the “right answer”.

In a more conventionally structured electricity system, long-term planning can be a very useful guide to policy and ministerial determinations of what mix of generation technologies might be licensed. But in South Africa, saddled as it is with a debt-laden Eskom, the whole exercise is much less useful. You can run one of the best electricity procurement processes in the world, as the IPP Office does, but if the participants can’t get connected to Eskom’s grid then it does not go anywhere.

In the same way that where countries that have fused the governing party and state itself, you don’t look to the legislatures for what is likely to happen but rather to the leadership of the governing party itself, a better place to figure out what lies in store for our electricity future lies within Eskom itself. We get clues from the Multi-Year Price Determination Process run by the national energy regulator, Nersa, and Eskom’s own disclosures, including those it chooses to make in its annual financial reporting. What passes for electricity planning in this country is really the outcome of a select coterie of analysts within Eskom who develop all the relevant investment plans.

Despite numerous failures in the past including the stupendous cost overruns and delays of Eskom’s new build programmes, persistent load shedding, and huge price increases, the standard of disclosure we get from Eskom is entirely inadequate. As such, the rest of the country suffers from a severe problem of information asymmetry. Eskom also has a monopoly on the information on its operations and that information determines the important investment choices.

Nersa has been able to rein in some of Eskom’s demands as it did in the present MYPD3 process and Eskom’s ill-fated effort last year to secure higher tariffs though a Regulatory Clearing Account (RCA) application. Yet, Eskom’s planners have been able to circumvent Nersa by going to Cabinet directly and more recently, Nersa itself has been depleted of the necessary skills and its powers are under threat – proposed changes may allow Eskom to appeal/review decisions that go against it directly to the politicians.

Without delving into the detail of Eskom’s disclosures, we can be sure that they are much less than the full truth. Why? Well, Eskom has to run two separate and incompatible stories concurrently. To Nersa, it is full of dark warnings about the catastrophic outcomes should Nersa not grant its MYPD tariff applications, but to National Treasury and its debt holders in its annual financial reporting, everything is just fine, management is of a superb quality and it is about to turn the corner into a glorious future.

Eskom is due to launch a fresh MYPD application so we have to expect a different story to the one we have just heard with the release of its 2016 annual financials. The basis upon which Eskom secures its average selling tariff in terms of the MYPD is pretty sweet.

It is a reasonably complex process but determination works roughly on a two-stage basis. The first and biggest determinant is where the value of Eskom’s asset base, known as the Regulated Asset Base (RAB), is calculated. This is not the same as set out in Eskom’s balance sheet. Depreciation works differently in the RAB. Using a delivery van as an analogous example, regular accounting might depreciate the van to zero in five years but in reality, the van still has a value which can be confirmed by the second-hand market. The RAB tries to determine the real value based on the theoretical replacement cost of, say, a 40-year-old generator (obviously there is no second-hand market for these). Thereafter a permitted return on this RAB is allowed. The level of the permitted return makes a huge difference and is always the subject of considerable dispute.

The second phase is based on the projected costs of serving an estimated amount of electricity over the next five years. These include fuel costs, salary overheads, operations and so forth. The measure used is “a prudent operator” (a tough decision as there are no other operators with which to compare Eskom). The outcome of the two calculations are added together and then divided by the predicted amount of electricity sold in each of the following five years to get a per kWh average tariff. Different customers then have their separate tariffs divvied out around this average amount. If Eskom’s sales are less than expected or fuel is higher, then Eskom can come back with an RCA application and get the shortfall filled in by securing a higher tariff for the following year. What this means is that Eskom has a strong incentive to overestimate demand since it can recover the revenue shortfall if its estimates are larger than actual electricity demand.

Eskom in this set-up has no market or capital risk and its mistakes in misreading demand or fuel costs are simply picked up by electricity consumers through the RCA. As we have seen, it is not working out for electricity consumers. Worse, the principal argument regulating Eskom’s tariff in this way does not exist any more. The justification for placing all the risk on consumers is purposefully to make Eskom a risk-free prospect for its lenders. A utility shielded from risk makes it an attractive borrower in the market and therefore should allow Eskom’s cost of borrowing to be ultra-low.

This, in turn, ought to allow Eskom to pass through the benefit of low borrowing costs for its capital projects into lower electricity tariffs. But this is far from what is actually happening. Eskom’s debt is “junk status” and its borrowing costs have ballooned to being 4% higher and more than government debt. The market doesn’t agree that Eskom is managed at all well.

This assessment is not without good reason. It is clear that Eskom has little idea of how to manage its capital expenditure or how to budget for it. Eskom’s most recent five-year capital expenditure plan budgets a total capital expenditure of R339-billion. To fund this, Eskom plans to raise additional debt of R68-billion a year for the first four years and another R53-billion in 2021. The budgeted capital expenditure is already significantly more than Eskom’s previous estimates of around R280-billion. The disastrous new-builds at Medupi and Kusile might account for the differences. Energy analyst Chris Yelland makes the point that the estimated all-in costs for the completion of Medupi and Kusile are at R208.7-billion and R239.4-billion up from the 2014 estimate of R154.2-billion and R172.2-billion respectively (an increase of 35% and 39%). We should reconcile ourselves to the fact that Eskom has no idea of what the cost of finalising Medupi and Kusile will be.

A recent RMB analysis says that the (current) budgeted R339-billion will be split with roughly R155.6-billion on the new-builds and R65.7-billion on technical work on the existing ?eet, distribution will get R64.4-billion and the transmission grid (IPPs and grid compliance); followed by other projects in transmission; and with some part going to electrification and fuel procurement (investing in coal mines). The problem is that even this is well short of what Eskom has said it needs. Eskom has previously said that just getting the transmission grid to N1 standard will require R163-billion on its own and according to Matshela Koko, Eskom’s group executive for generation, Eskom would need to find an additional R38-billion in the next five years to further capitalise the cost-plus mines.

Eskom’s confusion about its own business comes in an extraordinary statement in its 2016 integrated report (see page 91). In it, Eskom claims that IPP purchases account for 18% of primary energy, although IPPs contribute only 4% of electricity generated. Really? IPPs, such as the renewable energy plants, provide electricity from independent power generators. Eskom spent the equivalent of about 28c/kWh on coal across all its coal-fired plants but “forgets” to mention its capital component that still has to be added back. In short, Eskom tries to compare the cost of a taxi ride (driven, paid for and fuelled by someone else) with the cost of only the fuel needed when driving one’s own car. The facts of the matter are this: If Medupi and Kusile were owned independently, their kWh cost to Eskom would be way, way higher than the most recent renewable projects. If this is how Eskom understands how IPPs work then the prospects of a more diverse electricity system with more and cheaper renewables looks decidedly dim.

Other than multilateral and development funding institutions (which have high levels of conditionality), Eskom is probably not able to raise more debt in foreign markets at reasonable rates. It is lucky that it can still raise a lot of debt locally on our bond markets but the underlying reason for being able to raise the necessary debt is because of government support. We have seen several bailouts (accounting for most of the improvement in Eskom’s financial metrics) already and continued government support will be critical if Eskom is to raise the debt it needs. By 2021, Eskom’s total debt could be as high as R500-billion or as much as 20% of total government debt.

The depressing reality is that Eskom has the rest of the country, including National Treasury and Nersa, over a barrel. It presents a grave systemic threat to South Africa’s future prosperity. There are several options to try to reduce the threat Eskom poses. For one, its board of directors and who sits there needs careful reconsideration. Top management should also be scrutinised. It is completely unacceptable to have any conflict of interest or shortfalls in management competence. For example, nobody has replaced Caroline Henry as the head of Eskom’s treasury division and this role has simply been allocated to the financial director, Anoj Singh. As we have seen at SAA, the people generally required to do due diligence on funding are an impediment to corrupt practices. Other than at operational level, all bonuses should be based on Eskom improving its creditworthiness and on the amount of private capital through IPPs or otherwise is invested in the electricity system.

It is too late to change the regulatory methodology used to fix tariffs but what we ought to demand is far greater transparency to address our long-standing information asymmetry problem. South Africans should not accept claims of confidential or strategic information – there is no reason for confidentiality. Further, we should be entitled to understand the costs right down to each power station and its coal supply agreements, and we will also need to have accurate information on the capital programme with regular updates. What we can’t allow is having Eskom reallocate important projects in transmission and distribution down the line to the new-build programme simply because those have suffered yet further increases. Doing so comes at a cost of any IPP projects. Finally, we all need to understand exactly the way in which Eskom’s time of use tariffs are compiled.

We should be clear about the fact that the interests of the country are now quite separate from the interests of the company and this can’t continue. DM

Photo by Gavin Fordham.