Just find the wretched money for the diesel!

The economy is reeling from near-constant load shedding. These rotational blackouts are an enormous constraint on growth, and businesses (and households) are being forced to incur extra costs to keep productive.

In some cases alternative supply is largely unfeasible, resulting in things like significant production declines at the country’s largest steelmaker, similar issues at our mines, and even challenges with chicken supply which led to the forced closure of 70 KFC outlets last month.

For small businesses, the costs are even more severe.

And this disregards the lost output because of unnecessary traffic jams and so on.

It’s the inaction of government that’s most infuriating.

There is no plan.

All we have is short-termism. The last ‘intervention’ announced by government was the 50 million litres of diesel sourced by Public Enterprises Minister Pravin Gordhan from PetroSA in late November. It remains unknown how this was actually paid for. That stock lasted about a month (quite where Eskom is sourcing the diesel to burn – as recently as Tuesday night – is a mystery).

Eskom’s open cycle gas turbines (OCGTs) can generate up to 2 400MW of power, with those operated by independent power producers (IPPs) at Avon and Dedisa Peaking Power plants able to generate an additional 1 000MW.

They have a simple design – you burn diesel (ideally gas) and the turbine produces electricity, meaning that breakdowns are very rare.

However, running these constantly at full load is not possible due to the physical constraints of getting diesel to these plants.

Eskom’s own calculations state that this limit is R2.4 billion worth of diesel a month (at November prices); roughly 100 million litres. Never mind the fact that Eskom still pays R2.18 per litre – around 10% – in the form of the Road Accident Fund (RAF) levy for this fuel.

In round numbers, Eskom cannot burn much more than R25 billion* worth of diesel in a year at its OCGTs (and probably a further R10 billion at those operated by IPPs). Its own budget for diesel in the current fiscal was R6.1 billion; actual usage has been more than double that.

National Treasury all but shot down a request by the utility for an additional R19.5 billion late last year, with Finance Minister Enoch Godongwana telling Bloomberg that government doesn’t have the cash.

Still, these plants can reduce the severity of load shedding by about two stages, meaning that Stage 6 would be Stage 4 and Stage 4 would be Stage 2.

And with the fragile and unpredictable coal fleet, Eskom is being forced to artificially run a larger reserve operating margin during the evening peak than normal, precisely because it cannot rely on its diesel peaking plants in the case of multiple trips at baseload power stations.

In practical terms, since the diesel ‘ran out’ in December, load shedding during the evening peak is running at nearly one stage above the actual generation shortfall due to this. So we have Stage 4 load shedding instead of Stage 3, for instance.

It’s a no-brainer

The calculation on whether government ought to find the extra money needed for diesel is disarmingly simple.

Economists put the cost to the economy of Stage 6 load shedding during business hours at R500 million per hour.

Other, lower estimates have this cost at R50 million per stage per hour.

Lessening the acuteness of load shedding by two stages would therefore mitigate the damage to the economy by about R200 million.

We’d need just two working days of constant load shedding to justify the cost. Even if one were generous and halved the impact, you’d need less than a week to justify spending money on diesel.

(One particularly aggressive estimate puts the total cost to the economy last year at over R1 trillion.)

Any rational person would choose burning as much diesel as possible in the short- to medium term.

Many ways to skin a cat …

One can, on the other hand, understand Treasury’s reluctance in handing Eskom a pile of money to incinerate. There are multiple ways out of this crisis, though.

What if the peaking power plants were “bought” from Eskom and injected into a state-owned special-purpose vehicle (SPV) that is explicitly not allowed to make a profit?

These plants can then be operated separately from Eskom (and its problematic balance sheet). The SPV’s job would be to simply run the plants when the system operator says it must.

Or run this on a for-profit basis and involve the private sector. Establish this SPV as a public-private partnership (51% owned by government) with set – and limited – returns on the investment. Getting rid of the RAF levy and any other unnecessary levies on diesel will more than ‘pay’ for the returns (and the bulk of the construction).

An acceleration of the conversion of these peaking plants to also operate on natural gas is long overdue. Not only is gas more efficient, its carbon impact is significantly less than that of diesel.

A move to dual fuel plants will remove the bulk of the physical fuel constraints that exist today. Then the plants can run almost all month if need be.

Eskom has been “moving ahead” with the conversion of these plants to ones that can run on both diesel and gas for nearly a decade now. Just get on with it!

Why aren’t we building more OCGTs?

In 2004, Eskom under the leadership of Thulani Gcabashe, identified that it would not be able to generate enough electricity to meet the winter demand peaks in 2007.

OCGTs were selected to meet this shortfall, as their construction lead times were a fraction of that of coal or nuclear (under three years, versus more than 10).

The Ankerlig and Gourikwa stations, at Atlantis and Mossel Bay respectively, were built and ready for winter three years later. This is amazing in hindsight (especially when compared with Eskom’s transformation into an entity that is now entirely reactive).

Their combined cost? R3.5 billion.

In today’s money, that’s still under R10 billion.

Why aren’t we building more of these plants (instead of pinning our hopes on temporary Turkish power ships that we do not own and that may actually never arrive)? They’re almost laughably cheap.

Beyond the current emergency, we will still need peaking plants as we transition to renewables.

The sun doesn’t shine after 18:00 and the system operator will also need to be able to draw on resources that can be deployed instantly when there are sudden changes to generation (coal units trip, winds change, storms interrupt solar supply).

Eskom doesn’t have the balance sheet – nor, arguably, the skills – to build any new power plants, not even simple OCGT ones. Carving these plants out of Eskom will resolve this. (Minister of Mineral Resources and Energy Gwede Mantashe could even realise his dream of ‘Eskom II’!)

Of course, tariffs will need to change to reflect the higher costs of generating power during the morning and evening peaks.

Eskom’s 2022 results show that the cost per megawatt hour at OCGTs is about 10 times that of coal – up to R4 700/MWh vs R424/MWh.

Read: Eskom’s fifth straight loss compounds SA’s load shedding woes

The process to address the tariff mismatch is already underway and will incentivise industry, companies and households to use less power during these times (and to install their own storage solutions). This is the reality the world over.

Instead, we have Eskom shrugging and saying ‘we needed 4 000MW to 6 000MW of additional capacity yesterday’, a failed renewables bid window (Reipppp 6), a misguided resolution from the ruling party to move Eskom to the Department of Mineral Resources and Energy …

Oh, and inaction.

*The diesel price has dropped since November

Source: moneyweb.co.za