Johan Fourie's blog

I'd rather be a comma than a fullstop

High-speed rail in South Africa: too costly to consider

with 13 comments

On Thursday, Business Day reported that “the Department of Transport will commission its first independent, nationwide study into the cost and feasibility of high-speed trains between major cities, running at speeds as high as 400km/h”.

“The future of rail is high speed,” Mr Vilana said. In the “intermediate” term, the focus of the government, through state-owned commuter rail company the Passenger Rail Agency of South Africa (Prasa), was on getting “higher speeds” of between 140km/h and 160km/h on the narrow-gauge infrastructure.

“Certainly, going forward, we need to move to high-speed, standard-gauge rail,” said Prasa strategic network planning GM Hishaam Emeran.

For high-speed rail to succeed it must offer a “competitive journey time of about four hours over distances of 800km-1,000km”, he said, after which the trains would start to compete with airlines.

In the discussions on high-speed trains, “Durban to Johannesburg features strongly”, he said, adding that Cape Town to Johannesburg was also being mooted, “but that is a bit on long side … (unless) you start building key intermediate stops such as Kimberley or Bloemfontein and it starts opening up — you get a different picture then”.

There was also “huge demand from Polokwane to Gauteng, and then even further north. The feasibility studies will indicate where that is necessary,” he said.

High-speed rail is not a new idea, and it is enticing: it could provide South Africans and tourists with an affordable and safe way to travel long-distance, could reduce our dependence on oil, and, much like the 2010 FIFA World Cup and the Gautrain, could signal South Africa’s rise as an emerging power on the continent. Also, other countries are doing it: it’s difficult to imagine travelling in Europe without using the high-speed rails that permeate the landscape. But high-speed rail is not just a developed-country phenomenon any more. China has a vast rail network that is continually expanding: according to ChinaTimes, the country has recently completed “a 2298 kilometer line between the capital and Guangzhou which will cut the travel time between the two cities from 22 hours to just eight, with bullet trains travelling at 300 km per hour. The journey will take passengers through six provinces, 28 cities, 35 stops in major cities such as Zhenzhou, Wuhan and Chengsha, providing rapid rail service to 400 million people in the country’s heartland.” Many other developers, from Turkey, to Taiwan, to Turkmenistan, have high-speed trains.

Economic theory suggests that infrastructure is a key building block of economic growth: it facilitates lower transport costs, boosting trade and market integration. (Read a longish overview here.) Several economic history papers also show the significant long-term implications of rail infrastructure: Dan Bogart proves that transport revolutions before and during the Industrial Revolution decreased freight charges by 95 percent in real terms from 1700 to 1870 implying an annual TFP of more than 2 percent, meaning that “transport improvements were major factor in raising the standard of living in Britain and were as significant as other innovations”. Alex Moradi and Remi Jedwab show how two railway lines built in Ghana between 1901 and 1923 unintendedly opened vast expanses of tropical forest to cocoa cultivation, allowing Ghana to become the world’s largest producer. And in a recent essay, Bantu Mahali, a graduate student at Stellenbosch, shows how Free State farmers in the 1880s successfully agitated the colonial government to remap the proposed railway line to the mines in the interior to avoid passing through Basotho territory, then the bread-basket of the rapidly growing mining settlements. The result was that affluent Basotho farmers lost huge market share to cheap imports from the United States and Australia from where grain could be transported at a much lower cost to the mines (a distance of more than 8000 km!) than the grain grown on the slopes of the Drakensberg mountains just 300 km away. Incidentally, because of the persistence of railway lines, those spatial patterns still exist today.

High-speed rail seems to be the panacea to our problems. So I’ve let my imagination run wild and mapped what the ideal high-speed rail system would look like: Stage 1 would see a connection between Durban and Johannesburg, and between Johannesburg and Louis Trichardt (the North-South corridor). Stage 2 would extent this network to East (Maputo through Nelspruit) and West (to Gaborone through Rustenburg, and to Bloemfontein through Welkom). Stage 3 could see Bloemfontein extend it’s network to Kimberley and Maseru. Stage 4 would then link Bloemfontein to Cape Town through the arid Karoo. Further stages could be added extending north into Africa.


But, in truth, this can be no more than a pipe dream. The reason: it is simply too expensive. China currently builds the world’s cheapest high-speed rail, at $24 million per kilometre of line. (Germany is double that, Korea slightly lower at $40 million.) A railway line between Johannesburg and Durban would be about R120 billion (assuming a distance of 500 km and an exchange rate of R10/$), less than the R160 billion touted by Japan International Consultants, but still significantly more than the Medupi power plant that is currently under construction. If all the links on my map above were to be built, the (very conservative) costs would total R1 128 000 000 000 (or R1.1 trillion). That would double South Africa’s current external debt.

And these are just the construction costs. Land has to purchased to allow the new rails to be built (remember, South Africa’s existing railway lines are all smaller-gauge, which means they will have to be replaced entirely by standard-gauge on which these high-speed trains run). Land is probably cheap in the Karoo, but less so in the urban centres. And then there are the operational costs: these will hopefully be covered by user fees, but who will pay if they are not? In an earlier study, Estian Calitz and I looked at the financing options for the Gautrain and other such infrastructure projects. The construction of the Gautrain was financed by government (i.e. South Africa’s current and future tax payers), while the operational costs are paid for through user charges (tickets). The same model would apply to high-speed rail, but it is unclear whether there will be enough users to even cover operational costs. A plan to build a high-speed rail between Sydney and Melbourne in Australia was recently abandoned for these exact reasons.

A South African high-speed rail network sounds like a great idea. And there’s no doubt many will benefit from the greater freedom that low-cost travel brings. But, unfortunately, the cost of construction is simply too great, the opportunity cost simply too high. High-speed rail cannot be an option, at least for the foreseeable future.


Written by Johan Fourie

June 2, 2013 at 09:57

13 Responses

Subscribe to comments with RSS.

  1. … I’d thought it was all forgotten about these rail projects, but jus few days back i read an article that the Dbn-Jhb rail-link will materialise sometime in 2017 together with a rail link between Shaka Airport & Durban city centre.
    ….also I think there ws an agreement with the Chinese Dev Bank and SA government to finance construction, the figures were $30bn(US) from the Chinese, though not enough for all those projects but Im positive the Dbn Jo’burg link will materialise!


    June 4, 2015 at 04:04

    • Has there been any movement on the this project?


      March 3, 2017 at 12:42

  2. … I’d thought it was all forgotten about these rail projects, but jus few days back i read an article that the Dbn-Jhb rail-link will materialise sometime in 2017 together with a rail link between Shaka Airport & Durban city centre.
    ….also I think there ws an agreement with the Chinese Dev Bank and SA government to finance construction, the figures were $30bn(US) from the Chinese, though not enough for all those projects but Im positive the Dbn Jo’burg link will materialise.


    June 4, 2015 at 03:59

  3. I have a national diploma in Analytical chemistry and I would like to change to chemical engineering. Is it a wise choice? Is it possible that I do Birch in chemistry Eng from N.dip Analytical chemistry.


    May 24, 2015 at 19:02

  4. Perhaps there is a way of building those high-speed trains in narrow gauge. Even if they have to run at a reduced speed of 150 – 200kph it may be worth it.

    Barry Bilewitz

    June 27, 2014 at 05:52

  5. could u please provide me with d career options if u a doing cormecial subject which are accounting,business studies,economics and maths


    January 18, 2014 at 05:38

  6. The cost is not upfront it is spread out through the duration of the project, i.e the Johannesburg/Durban phase could take up to 8 yeaars to plan/design/construct/commssion at that time employment would be created not just in the public but also private sector (including all the supplychains and other sectors), which brings in growth i.e (tax revenues increase). The UK is pursuing with high speed rail because in the long term it brings in jobs, opportunity and growth in the economy, which SA desperately needs at this time!

    Sam China

    September 23, 2013 at 17:06

  7. Cool share on High-speed rail in South Africa: too costly to consider | Johan Fourie’s blog.

    It is surely among the most helpful that Ive read in quite a while.

    web design Cape Town

    September 18, 2013 at 20:31

  8. […] here’s an idea, South Africa: instead of building an expensive (as I’ve suggested in an earlier post) high-speed rail network, why not give Musk R100 billion to build his Hyperloop in South Africa. […]

  9. Loved your article – one idea to pursue further however might be the idea of ‘inter-city’ high speed trains. For instance, a bullet train from Khayelitsha to Cape Town CBD, using the figures from your article a guess-estimate of the cost would be +/- 7.5 billion ZAR, but if it hit an average speed of 250km/h, it could seemingly make the trip in 8 minutes… crazy expensive, but might be one way to ‘compress’ the city, an unusual but bold way to combat sprawl?

    Stuart Paul Denoon-Stevens

    July 2, 2013 at 01:41

  10. The only reason I pick from your report for the project to fail in South Africa is that it failed in Australia. You could have kept quite on the costs if what you wrote up in the document was all you had to say.


    June 4, 2013 at 15:35

  11. Reblogged by SA BLOGGERS –


    June 3, 2013 at 15:21

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: