Eskom’s Majuba power station rail project debacle

Another a stark reminder of the massive costs when engineering projects run out of control and policy initiatives have debilitating unintended consequences.
The policy of trucking coal to Majuba has created a sense of entitlement both within the utility and externally (2014 file photo of Majuba rail section near Ermelo). Image: Supplied

The original business case for the 68km Majuba Rail Project was authorised in December 2004 with a board-approved cost to completion of R1.5 billion. This has since ballooned to an estimated R5.9 billion currently, plus R2.1 billion in capitalised interest during construction. The project is running many years late, and is now only expected to be completed by the end of March 2021.

Besides the above project capital cost overruns, the extended time overruns have incurred significant additional internal operating costs and external road repair costs (see below) that have further added to the devastating impact this project has had on Eskom, the built environment, and ultimately on the economy.

Nobody at Eskom appears particularly interested in quantifying these openly, perhaps because the realities are simply to ghastly to share publicly.

Background to the project

Eskom’s 4 110 megawatt (MW) Majuba coal-fired power station, situated near the towns of Amersfoort and Ermelo in the Gert Sibande District Municipality of Mpumalanga, comprises three 657MW dry air-cooled units and three 712MW water-cooled units. Construction on the power station started in September 1983, and by April 1996, 13 years later, only the first unit was connected to the grid. The last unit was commissioned in April 2001.

Fourteen million tons of coal per annum (14Mtpa) is required to fuel the power station, and this was supposed to come from a new tied-colliery built and operated by Rand Mines. However, development of the underground works stopped in 1993, and the mine was prematurely decommissioned when geological faults were belatedly discovered in the coal seam.

As a result of the decommissioning of Majuba Colliery, a branch rail link with a capacity of 8Mtpa was built from Palmford on the Natal Corridor (Natcor) general freight line between Standerton and Volksrust to Majuba Power Station. Due to a surplus of Eskom generation capacity at that time, the construction of Majuba was deliberately delayed, and at the end of the second quarter of 1996, only the first unit was in commercial operation.

The capacity of the Palmford rail link from the Natcor line was therefore more than adequate to meet the initial coal supply requirements of Majuba Power Station.

But from about 2003/2004, Eskom commenced a policy of procuring coal from new emerging sources, which resulted in delivery of coal by road even though the capacity of the Palmford line could still accommodate the volume of coal required. This was because the location of some of the new coal sources contracted to supply coal to Majuba made it impossible to move coal to Majuba via the Palmford line, and coal deliveries by rail were thus supplemented and displaced by delivery of coal by road.

The Majuba Coal Transportation System (CTS) Project

However, based on forecasts of electricity demand growth, Majuba Power Station had to gear up from a burn rate of 2.2Mtpa in early 2000 to a required capacity of 14Mtpa by 2010. The Palmford/Natcor line could not accommodate the increase in volumes, and in early 2000/2001 the need was identified to build a 68km rail line from Ermelo to Majuba, linking onto the heavy haul export rail line from Mpumalanga to Richards Bay.

The new line would have sufficient capacity to transport the full 14Mtpa of coal required by Majuba, which was to be offloaded at the Majuba terminal via an upgraded coal tippler system. This link would also shorten the distance that coal would have to be transported by rail from the Witbank coalfields, thus reducing rail coal supply costs by an estimated 15%. The original business case for this CTS project was approved by the Eskom board in December 2004 with a cost to completion of R1.5 billion.

Challenges in executing the project

However, significant “challenges” have been experienced by Eskom in executing the project.

These include delays in land acquisition for the rail servitude, funding constraints (which finally resulted in a R6 billion allocation in a $3.75 billion loan to Eskom by the World Bank in 2010), unforeseen underground conditions, adverse weather conditions, scope changes and disruptions by community pressure groups. The inquiry for the coal stockyard optimisation had to be issued three times due to unresponsiveness by the market. The works required for the interface with operational plant was delayed due to the need for power station outages. These outages were deferred several times due to issues with low coal stock levels at the power station. There were also problems with the signalling system contract and signalling equipment.

Finally, the main contractor went into business rescue and cancelled the contract with Eskom in July 2019, which led to subcontractors also leaving site.

In December 2019, an incident occurred at the power station where the coal tippler takeout conveyor caught fire, and all coal deliveries by rail to Majuba since that date have ceased.

This means that since December 2019, the full coal supply requirements of Majuba Power Station have been met by road transport.

There have since been delays by National Treasury in the adjudication of a single-source contract – which is critical for the implementation of an expedited execution of the recovery works – and the completion of Majuba Rail Project, which is currently estimated by Eskom to be 87% complete.

On top of all this, Eskom has acknowledged that its own management of the project has been lacking. There have been several different project managers during the execution of the project, with varying qualifications and experience.

While Eskom has declined to name or discuss individual performance or names, it is in the public domain that poor project management is at the heart of a dispute between Eskom chief operating officer Jan Oberholzer and Mark Chettiar, the former general manager: coal and clean technology, and project manager for the Majuba Rail Project and the new ash dam project at Camden.

When confronted by Oberholzer about alleged underperformance in managing his project portfolio in September 2019, Chettiar informed the COO that he was taking “sick leave”, and has since been absent on full pay while he pursues the matter at the Commission for Conciliation, Mediation and Arbitration.

Chettiar has denied all wrongdoing.

Additional internal operating costs and external road repair costs

When responding to specific questions on the impacts of fraud and under-delivery in the volumes and quality of coal to Majuba by road since 2003/2004, Eskom responded that there have been no such impacts arising from this.

However, various internal and external investigations and reports have indicated that since 2003/2004, the diversification of the coal supply within Eskom resulted in something of a free-for-all, where nepotism and fraud were apparent in the award of coal supply and transportation contracts to friends, family members and other politically connected persons.

Eskom has also indicated previously that the most frequent instances of coal supply fraud and coal quality breaches occur during the transport of coal by road by unscrupulous contractors to the various coal-fired power stations in South Africa, and that satellite monitoring and surveillance technologies have confirmed this.

Finally, the first tender process by Eskom since 2003/2004 for the transport of coal by road to Majuba Power Station was conducted in 2019, and the utility has since advised that this resulted in a 30% reduction in road transport costs to Eskom. From this, a good estimate of the massive operating losses suffered by Eskom at Majuba through excessive road transportation costs alone over the last 15 years can be made.

Furthermore, currently, at Majuba alone, Eskom is incurring additional costs of R62 million per year as a result of road transport, over and above what it would have paid for the transport of coal to Majuba using the old Palmford/Natcor rail link.

In addition, Eskom has indicated that had the new Majuba rail link been completed as planned, it could handle the full 14Mtpa coal requirements of Majuba, and would reduce the cost of coal to Majuba by 15% compared to that supplied via the old Palmford/Natcor rail link. Again, from this, a good estimate can be made of the massive operating losses suffered by Eskom over the last decade as a result of the many years of delay in the completion of the Majuba Rail Project.

Finally, there are the external costs of wear, tear and repair of local, provincial and national roads arising from the transport of coal by road to Majuba Power Station.

When asked about these road repair and resurfacing costs since 2003/2004, eyes glaze over and things become quiet and vague. It seems nobody at Eskom is prepared to venture a guess.

However, a competent firm of civil consulting engineers with road design experience could easily perform a study and provide a good estimate.

But there are now so many vested interests along the road transport supply chain, both within Eskom and in local communities and businesses, that nobody appears particularly interested in quantifying the cost of externalities outside of the direct cost to completion and capitalised interest during construction of the Majuba Rail Project. Perhaps the figures are too disturbing to the policy positions and project decisions already taken.

Eskom now acknowledges that even after the completion of the Majuba Rail Project, road transportation will continue for some time, and in fact, will probably never stop.


The policy started in 2003/2004 to truck coal to Majuba has spawned an entire ecosystem that feeds off the largesse of a state-owned enterprise that was out of control. A sense of entitlement has developed among parasitic elements – both within Eskom and externally – to extract as much they can, with little sense as to value for money and efficiency, to the wellbeing and sustainability of the host body, or to the national interest.

This also begs the question as to whether the extended project delays on the Majuba Rail Project may have been deliberately engineered by individuals, internal and external to Eskom, whose revenue streams – and those of their politically connected associates, friends and family members in the local coal mining and road transport businesses – would be negatively impacted by the timely completion of the project.

The fire incident in December 2019 that stopped all deliveries of coal by rail to Majuba Power Station is a particular case in point.


Sort by:
  • Oldest first
  • Newest first
  • Top voted

You must be signed in and an Insider Gold subscriber to comment.


The photo says it all.

Construction methods are maybe not quite the best.

The scaffolding would cost more than the pillar. Oh. Maybe the skills have been lost??? Maybe the people that knows how to build those things are long gone and the one’s that are still around have found something else to do.

Go for it! I cant say what I want as I will be “Moderated” but you know what I mean.

JUST LOOK AT THE PHOTO????? $%^&**&^%$%^&*T SAKES

I understand your irritation. They can do it in a more cost-effective way, but in typical Gosplan fashion, they want to save and create jobs. The cost incurred with these inefficiencies is ten times more than the wages. They would be more efficient if they paid the union members to stay at home and let competent people complete the project. The ANC fails to realise that if they follow Gosplan strategies, they will reap Gosplan results.

I am not familiar with scaffolding – please explain what I should look for,and understand, in the photograph! I’m looking for education not an argument.

The whole country needs another JVR ?

I have an instant headache!

if government believes it should own and manage places like Eskom it should insist all its major capex contracts are turnkey. Eskom has NEVER – go back 70 years if you want – delivered any project on time on budget

Voting ANC has ‘debilitating and unintended consequences’.

Why are we not surprised ?

This article picks up some very valid points. However, it is sadly just the very tip of the iceberg.

We have been intricately involved in these attempts by the good(yes there still are some) powers that be to switch the ESKOM system from road to rail. Serious attempts started to be made in 2007 when just no headway was being made. This actually accelerated in the 2008/9 commodity crash as capacity was freed up. What i can say is limited from a moderation perspective and also that some of this is still part of ongoing investigations and NDAs but at a vastly diminished level. We are not the only ones, but most have been chased away. The author is so spot on with regards vested interests. There is no interest in allowing this to go to rail – the money made on road is obscene as are the comparative and relative costs. Those of you with knowledge in logistics will understand the economies of scale achieved when one train pulling 6,000 tons is delivered, compared to the same quantity being delivered on 200 x separate 34ton side tipper links.

Majuba consumes 14mtpa. That equates to approx 5 trains per day. When delivered by road it is the equivalent to 1,300 trucks per day.

Power stations are designed to be fed by rail, first and foremost – road should only ever be used in emergency situations, such as serious derailments, or other adverse conditions such as excessive rains or mine supply issues. The original designs were always 90/10 (rail/road) – in a vast number of cases this became totally perverted and reversed, where in the case of Majuba now it is 100% – but many more power stations are constantly forced into very expensive majority road deliveries.

There is a figure in this article of R62m per year being paid over and above rail for road deliveries. Without advising exact numbers, all i will say is this is a fraction of the extra being paid. Let me rather put it this way. 14 million tons per year. Does anyone really think that the cost of moving all this by road would equate to ONLY R4.00/ ton extra!?

There are so many more moving parts to this article than the author is revealing… but that is likely for good reason. However, on round trips of 140kms, block rail will cost – even with TFR (Transnet Freight Rail) about 28% of that of road, it should be about 11.5%. Rail cost is around R33.00/ ton. Road cost is closer to R152.00/ ton. The difference is R119/ mt more for road – or R1.6 billion per year not R62m. Verify numbers yourself if you like you can verify on the net, this is no secret.

There is further digging which is required. Most road coal contracts are covered under wet contracts, meaning a lot of them the diesel is supplied. There is barely any control on this. We had issues over the years and detailed in several hundred reports, of trucks consuming over 100/even 200litres/ 100km, which is impossible – but trucking companies were literally running separate fuel businesses and/ or their drivers were – in many cases it also included the diesel suppliers and the diesel procurement divisions within Eskom. Power stations require continuous supplies 24/7/365 – meaning fleets of trucks run with multiple drivers. Allowing for maximum hours/ days off/ driver rotation/ sick leave/ go slows/ union action etc – each truck has 4 drivers allocated, sometimes this would be as many as 8 or 9 drivers. A lot of these drivers were and are ghost drivers. They either didnt exist, or if they did, they would be getting their salary effectively from Eskom but could very well be running fruit in the western cape. In this regard i am aware of large companies that have been created just supplying ‘contract drivers’ at what seems like cut throat prices. However, when Eskom is paying for several hundred drivers delivering coal at R17k a month basic – unscrupulous contractors can then offer drivers elsewhere in the country/ province at R6k/ driver / month. There are allegedly dozens of cases where sets of drivers are duplicated within the same organisation (Eskom), on different power stations and mines, in effect being paid for multiple times by the same company.

You get contracts from Eskom which you take to the banks and trailer manufacturers. You get finance easily on the back of these contracts to buy horses from Merc/ Scania/ Volvo etc. From Henred/ SA truck bodies/ Paramount etc etc. Some of these horses run once a month on a coal route then they are somewhere else – in a lot of instances no one checks. Huge finance is provided for trailers – the same story applies – these trailer companies are fine but they do not control where their trailers go. We have come across refrigerated trailers hauling coal, i kid you not.

The next level is where with this finance they use funds to purchase ghost trailers – there are a few builders out there that will go and obtain blocks of chassis numbers from the licencing department – and sign them off as produced and delivered – but they never are. So you can imagine a Wesbank as an example dolling out hundreds of millions to a manufacturer, who never delivers any equipment – just the pink slips to Wesbank, the cash is then split up cleverly between the interested parties. If random inspections are called for – the trailers are not available for inspection – if need be they will be produced if required. I can go on and on, there are so many levels and layers of corruption within just this one aspect of ESKOM and that is ROAD vs RAIL.

Road is also preferred by the coal ‘suppliers’ as they can very easily load up several hundred trucks with reject coal – which is nearly 75% soil – the supply chain is warned ahead of arrival and every 30 trucks tip on a different boiler pile, as such when this coal flows through the furnace on the belts it is mixed – for sure temperature and boiler pressures drop but usually not enough to kick them out. Road even means trucks can load from mines not even authorised as suppliers. Rail prevents this as stock piles are massive and pre-checked. Road is preferred as there is just no control, it is the wild west.

Think about the fact we are talking about one power station here – think of how many power stations there are, how many mines, suppliers. Eskom are paying tens of billions more per year on ROAD over RAIL and billions more on substandard coal as a result and the lost productivity that it brings.

I doubt it will surprise many here – that there are ex Eskom board members and ex Transnet board members that had and still have interests in road transport companies that transport coal into Eskom power plants. These are active investigations so no names. But did you ever wonder why RAIL always gets screwed up – a hint it is not by accident.

A report was issued in 2007 warning of massive power failures, as a result of logistics short falls and the growing problem of road and failure of rail. It was very easy to fix – yet we all remember 2009 when the country was devastated with our first real black outs. Yes there was equipment issues in Eskom but massive parts revolved around logistics failures.

The author alludes to something in the last paragraph, regarding the fire on the tippler. This is another investigation, let us just say it was likely not natural causes. The same reason TRANSNET tells Eskom they have no wagon capacity – is nonsense, they have a surplus of 11,000 x heavy bogey SMLJ 63 ton wagons, easily adapted to coal transport – yet these along with several thousand bulk coal ctrs stand rusting all over the country in obscure sidings. The question is why? Again the author makes it very clear.

Our group were eventually pushed off the ROAD to RAIL projects, along with all other suitors, through non payment and the obvious fact that there was and still is no drive from the very top of the political spectrum and from Eskom.

So in summary Eskom is in dire straits, however, a lot of problems amounting to several billion a month could be addressed literally overnight, but there is no will at the moment. Several thousand pages of reports detailing it all stand in boxes and on hard drives but there are those that never want them to see the light of day.

Reports have been done that actually drastically inflate the rail costs to make road look a lot more effective. Transnet then agree and the process goes back to the drawing boards and ROAD carries on – it is by far the biggest current and ongoing SOE fraud in existence.

R450B sounds scary and it is – but this could be rapidly decimated by just addressing the logistics. The Eskom operational problems are far from insurmountable but the political will is, for now.

Moderators if you want me to modify this no problem – i have not named names or companies and clearly your author is aware of a lot more than is in the article. I am not placing anything in the public domain which is not already known and has been for well over a decade.

Great reply. I would like to add the state of potholed roads that were never engineered for 30t trucks, especially not at the frequency of traffic.

Your comment should be published widely. Please forward it to a national newspaper or Daily Maverick!

Jan you are right but for now i think i would rather use my knowledge to help Eskom and RSA if i can. I also still choose to live here 🙂

Maybe someone reading this will remember my contributions/abilities and contact me again. I require neither recognition nor remuneration. Only one request, minimal interference to execute operations.

The South African of 2020 constantly mutters about incompetence and how “this would never have happened in the old days”.
Majuba power station started in 1983 and took 18 years to complete. The adjacent mine encountered a fault and had to be abandoned. The root issue here appears to be severe incompetence in the Eskom ranks in the 1980’s when this project was conceived and executed.

If you read the whole article you would know that the delay in construction of Majuba was deliberate due to a surplus of generation capacity at that time. Unlike today’s delayes at Eskom which are caused by mismanagement and corruption. During the 70/80’s Eskom built a number of large standardised “six-packs” coal-fired power plants that could produce power at very low cost due to the large economies of scale. Unfortunately the skills that made this possible are long gone now. However if government would allow the rapid roll out of IPPs our current energy crisis could be solved in the medium term, but currently the opposite is happening.

And this is while the rest of the world is moving on with their energy transition at an astronomical speed.
Wind generated electricity has fallen with 50-70%, PV solar 70-85% over the last 10 years.
Portugal again became the world record holder by beating or equaling a new record low for a PV project at US$ 1.35 cent/kWh. Abu Dhabi signed a contract at the beginning of the year for about the same price for a 2 G watt project. SA has better solar radiation than Portugal. The obvious problem of intermittency of renewable power production is solved with revolutionary pace.
Two contracts for large PV + battery storage projects were signed last year for locations near LA, Ca and in Nevada for 3-4 US cents/kWh, for power during the evening peak.
I for many years had already the idea that with the right combination of several renewable technologies located on a wide array of places, combined with a better HV-DC connected transmission system, could solve the intermittency issue for 60-80%. The rest could come from natural gas and Koeberg.
Now battery storage appears an ever more affordable and practical solution.
SA only gets about 6% of all electricity from renewables,and round 5 of REIPP(renewable energy independent power producers) contracts is still awaiting to be implemented for 3 or more years.
When the right size of projects can be implemented, prices should typically lie between 30 to 70 Randela cents/kWh. The whole BEE component and community participation part of these SA projects should just be completely scrapped to ensure speed and low price.
During their last north west European storm, more than 60% of all power in the UK was derived from wind.

The stench of ANC caders corruption and incompetence just is not going away.

Majuba should by all what is written here, by CY and Romulus, be one of the most expensive producers of powers.
Close it down ASAP !
Initiate Round 5 of REIPPs for 10 GW immediately. And remove all the unnecessary bureaucracy, BEE and community requirements from these energy projects. All private renewable power projects have been built, got up and running between 18-36 months, after all paperwork and licensing was done.

End of comments.



Subscribe to our mailing list

* indicates required
Moneyweb newsletters

Instrument Details  

You do not have any portfolios, please create one here.
You do not have an alert portfolio, please create one here.

Follow us:

Search Articles:
Click a Company: