Grant McDermott bio photo

Grant McDermott

Assistant Professor
Dept. of Economics
University of Oregon

Email Twitter LinkedIn   Scholar   ORCID GitHub
Very interesting:
South Africa is to unveil plans this week for what it claims will be the world's biggest solar power plant – a radical step in a coal-dependent country where one in six people still lacks electricity.
The project, expected to cost up to R200bn (£18.42bn), would aim by the end of its first decade to achieve an annual output of five gigawatts (GW)* of electricity - currently one-tenth of South Africa's energy needs.
Giant mirrors and solar panels would be spread across the Northern Cape province, which the government says is among the sunniest 3% of regions in the world with minimal cloud or rain.
Apparently the Dept. of Energy is going on a big investor drive to drum up support for the whole shebang. Apart from South Africa's Renewable Energy Feed-In Tariffs (REFIT) scheme - I'm not really in a position to make predictions about the economic implications at this stage... Well, apart from the rather obvious matter that anything requiring a subsidy is going to be more expensive than the current going rate. Still, until we get a carbon price up to internalise the externality costs of coal, I suppose subsidies will have to do to for kick-starting the broader integration of renewables in South Africa and beyond. (Lord knows we could also do with a bit of diversification in the electricity supply down there... And I won't even mention the need to do away with Eskom's crippling monopoly on power supply.)

General aversions to monopolies (and certain subsidies) aside, I've been pretty positive about CSP (concentrated solar power) in that part of the country for a while. One concern I do have about these big solar projects though, is that they tend to be in - well - hot and dry places. This becomes problematic given that you need pretty substantial amounts of water for cooling (with thermal plants) and even cleaning (for photovoltaics). I guess it was inevitable that they've got their eyes set on the Orange River for this one... I'm interested to see what the EIA throws up.

* Unfortunately, the article contains a frustratingly common mistake here. The author surely means "5GW of capacity" (not output) as the only meaningful way to interpret output is watt hours... It happens again later on when he writes: "South Africa currently consumes 45-48GW of power per year." (If you're interested, you can visit the reliable StatsSA to see that total electricity consumption in 2009 was the order of 230TWh, i.e. terrawatt hours.)