Analysis: The long goodbye to cheap and abundant electricity
- Dirk de Vos
- South Africa
- 09 Dec 2014 12:40 (South Africa)
Cheap and abundant supplies of electricity, all of it provided by state-owned Eskom. That has been the underlying assumption of almost all South Africans for over a generation. This assumption, over the long term, is a faulty one, and if South Africa is to get through its present crisis, then we will have to preserve and build our country’s economy using an entirely different foundation. It won’t be easy – for anyone. Cherished policies about our economy will not survive, and the way our country is managed will have to change too. By DIRK DE VOS.
The Minister of Public Enterprises, Lynne Brown, recently called upon South Africans to be patient with ongoing load-shedding until 2018, presumably the date by which she has been advised that Medupi and Kusile will become functionally operational. Once they are operational at 9,600MW, they will represent roughly 20% of Eskom’s generating capacity. On the other hand, Eskom’s creaking existing generating plant will need to be retired. But we need to be clear. The dramatic fall-off in Eskom’s plant availability is not only because of a lack of maintenance and running them too hard at the behest of politicians wanting to “keep the lights on at any cost”. It is also because its plant is getting old and creaky. Following the commissioning of Medupi and Kusile, then, we will quickly be right back to where we are now as far as generating capacity is concerned, but with one significant difference – our electricity will be far more expensive.
To understand this, let’s go back to the basic electricity utility business model. Any electricity utility sells its product, electricity, at a specific rate or against a tariff. This tariff, often reflected in the unit kilowatt hours (kWh) needs to be high enough to cover what is known as the levelised cost of electricity (LCOE) or the all-in cost of delivery of the electricity to the grid. As Eskom also manages the transmission and a big part of the distribution grid, its LCOE has to take into account the grid itself. There are three basic inputs that determine the LCOE: Capital cost to build the infrastructure (and the cost of that capital); the fuel costs and the Operations and Maintenance which include fixed overheads like salaries. About 92% of Eskom’s power is generated by coal-fired power stations (about 5% nuclear, 3% renewables) because South Africa has large reserves of coal.
One might assume that because we have large reserves of relatively cheap coal, located mostly near Eskom’s power stations, it means that we can have cheap electricity by right. Strictly speaking, this is not true. Large electricity generators are not off-the-shelf items but benchmarking studies are done on what electricity from different energy sources should cost. Recent studies, using a 10% discount rate (or the blended required return on equity and the interest rate for debt to finance the build) but excluding carbon taxes or carbon capture, estimates the contribution of the three basic inputs to the LOCE as follows: 57% to capital costs 32% to the costs of the fuel and 11% to operations and maintenance. On the same basis, nuclear’s capital costs represent 75% of its LCOE and renewables’ capital cost is closer to 90% (their fuel is free).
This is important because Medupi and Kusile will have cost at least double what they should have cost. The cost of building these plants, representing the input factor that is more important than all the others combined, means that that every unit of electricity generated by them will be expensive – this fate is now “baked in”.
Much higher tariffs will be a shock to our system. When Eskom was incorporated as a company in 2001, its current fleet of power stations had mostly been paid off and Eskom had very little debt. For two decades before that already, South Africa has been supplied with electricity by a utility that did not account for the costs of replacing its capital stock. So, Eskom’s tariffs have been less than half of what they should have been.
In the meantime, Eskom has developed a very expensive salary overhead, at about R623,000 per annum per employee. It is a useful exercise to compare the salaries of Eskom employees involved in distribution and their peers employed by municipalities to do the same thing – the differences are big. The salary cost overhead is one thing; Eskom is now up to its eyeballs in debt.
To be charitable, perhaps cheap electricity was a necessary component of the transition phase of our country’s history. Maybe very cheap electricity was the glue that was needed to hold it all together. It allowed significant energy intensive parts of the economy to continue to operate and we were able to extract from Eskom, a democratic dividend of expanded access to electrification and subsidised or free electricity. That path has now come to an end - it was inevitable.
Sadly, the overwhelming evidence is that our economy has not used cheap electricity well and at the same time we have become dependent on it. The South African economy shares many of the characteristics we see in so called “petro-states”. Our economy’s energy intensity is amongst the highest in the world. The amount of value created for a fixed amount of energy input is amongst the lowest in the world. Some activities like aluminium smelting and a large section of our gold mining industry, both foreign exchange earners, would not exist but for cheap electricity. We have avoided getting to grips with our real problems that prevent us being a competitive economy. Our labour productivity is very low and the best means to turn this around, a sound education system, seems as far away as ever.
Cheap electricity is also an important part of how we fund local government. The only part of our electricity system that is not run by Eskom is the approximately 70 distributors around the country – the municipalities. These municipalities sell approximately 40% of all the electricity sold in the country. The larger municipalities have just two sources of revenue: property taxes based on the value of the properties within their jurisdiction and service fees from the sale of electricity (water and waste removal are much smaller components). Electricity sales represent as much as 30% of total municipal revenues. Cheap electricity has been used as another tax by all local authorities and this source of revenue. When electricity was cheap, it was relatively easy to fund subsidised or free electricity. It also allowed some municipalities to get into bad habits by adopting a pretty casual approach to non-recovery and theft of electricity. That is no longer possible.
As one would expect, the recent increases in tariffs and now the power outages have reduced the amount of electricity being sold by all municipalities and this is now impacting on municipal finances. The trajectory of electricity consumption is now far lower than the lowest estimates in the 2013 update to the Integrated Resource Plan.
Like Eskom, local municipalities need to be part of the programme to have us conserve electricity while, at the same time, the sale of it supports their funding model. So, just as the finance minister is looking to raise new taxes to support his national budgets, you can be sure that local government is also looking to increase property taxes.
What to do, then? As the Chinese proverb goes, “The best time to plant a tree was 20 years ago - the second best time is now”, is instructive. We should have done what needed to be done 20 years ago to have avoided the jam we are in now, but what is done is done. Whatever decisions are made, cheap electricity is behind us, but we should make the decisions that have the best chance of making it as inexpensive as possible.
Getting stuck into a pointless debate about the merits of privatising Eskom takes things no further. But it should be obvious that Eskom should have no future role in building any further power stations. Eskom has high operational costs and because it is now so indebted, its cost of capital having recently been downgraded by Moody’s to sub-investment status (and being the most important factor in any new builds) is not particularly competitive.
If we are to have security of supply and a functional electricity system, these are the things that should be done:
- Develop a clear idea of what the LCOE of Medupi and Kusile as stand-alone projects will be. This is important because it provides a new benchmark against which any future energy projects should be measured. Various co-generation projects could be up and running quickly, but they have to sell into Eskom at tariffs that are too low - tariffs, as we have seen, that are not sustainable for Eskom itself.
- If the Independent Systems Market Operator bill can’t be passed, at least create an administrative split between Eskom’s power stations and the national grid.
- Together with Eskom’s Multi-Year Price Determination processes (they have a three-year duration) develop a 10-year view on what prices are likely to be and publish this so that everyone can re-adjust their own forecasts and investment decisions.
- Update the Integrated Resource Plan with the latest data of electricity consumption and do a thorough analysis of the impact of increased electricity prices on different sectors in the economy with a focus on job losses and wider economic growth forecasts. Remodel electricity demand and energy intensity for this new scenario. Expected growth in demand for capacity might well be below current demand.
- Expedite and expand all of the IPP programmes that have the capability of bringing new megawatts to the grid quickly.
- Re-do the national budget to take account of our changed cost of electricity regime. Importantly, make all electricity subsidies be part of the social welfare budget and adjust the overall budget including spending commitments for lower taxes from the economy. Focus more efforts on social cohesion. The next period of our history is going to be tough on everyone and the centre must hold.
- Insist that local governments re-adjust their own budgets so that, over time, the sale of electricity has a neutral impact. The path to this is via a mass deployment of time of use (or smart metering). Using telemetry rather than expensive roll-out of smart meters can be done today. As electricity becomes more expensive, people will use less of it but usage patterns become more “peaky”. Variable time-based billing smoothes this out. Charging for electricity will also need to change to having a line charge (to maintain the distribution network) and a consumption charge. We should then get municipalities to pursue energy savings programmes aggressively and feed in programmes that encourage rooftop solar and provide for feeding surplus electricity back into the grid.
- The aluminium smelters near Richards Bay and in Mozambique operating on extended Eskom contracts running to 2028 ought to be immediately acquired and then shuttered. They consume over 5% of Eskom’s full capacity and do so at a cost of less than 30c/kWh, less than half of Eskom’s average selling price of 74c/kWh.
- Eskom is running the Ankerlig gas fired Open Cycle power station near Atlantis as a base load power station, using diesel at enormous cost. This, too, is unsustainable. We need to fast-track the development of an LNG import and storage facilities at Saldanha Bay and convert Ankerlig to a Combined Cycle Gas facility (effectively adding turbo-chargers) using this imported LNG gas. The Kouga Port should consider the same thing. Mossel Bay can be serviced by rail. By removing coastal cities as importers of electricity, transmission losses of between 10%-20% resulting from getting Eskom’s electricity to those places could be saved.
- Look to the gas boom in Mozambique and procure either more gas or electricity generated from that gas for delivery to the Gauteng industrial heartland. For Mozambique to finance the development of its ample gas resources, it needs an off-take agreement. Concessionary development funding exists but it needs South Africa as the off-take customer. Let’s also agree that the transmission lines from Cahora Bassa be managed by a single entity on both sides of the border.
- Remove Eskom as South Africa’s representative to the Southern African Power Pool (SAPP) and replace it with the IPP Unit with a mandate to procure electricity from any other member of the Pool on non-discriminatory terms.
- Amend the Electricity Regulation Act, the regulations surrounding the Municipal Finance Act and the Municipal Management Finance Act to allow municipalities to procure to allow qualified municipalities to run their own independent power procurement processes. Many of our metros have a better credit rating than Eskom itself. The Renewable Energy programme run under the Department of Energy has done a great job in “finding” the price of different types of renewable energy - municipalities could run simplified feed-in tariff programmes based on defined capacity requirements.
The transition from an economy and society based on cheap electricity to one which is supplied with electricity at the price that it costs to produce will be a painful one full of tough decisions. It is self-evident that we can’t delay the process any further. It will need patience and fortitude – it will also need leadership. Failing to do anything about our crisis (yes, a crisis) is no longer an option. DM
Photo: Residents cook on a roadside during a routine power outage due to load shedding in Masiphumelele, Cape Town, South Africa, 07 December 2014. South Africa's parastatal Eskom is fighting a battle to prevent the possible collapse of the country's electricity grid. The energy supplier has implemented various stages of load shedding due to a lack of capacity as last resort to avert rolling blackouts which could bring the country to a standstill. EPA/NIC BOTHMA
Reader notice: Our comments service provider, Civil Comments, has stopped operating and will terminate services on 20th Dec 2017. As a result, we will be searching for another platform for our readers. We aim to have this done with the launch of our new site in early 2018 and apologise for the inconvenience.