DeLong Smackdown Watch: Tim Harford

Hoisted from Comments: Nicholas Weaver on Solar vs. Nuclear, Myhrvold, Dubner, and Levitt

Hoisted from Comments: Nicholas Weaver, who knows what he is talking about, writes:

The Very Last Superfreakonomics Post of All Time...: Rather than just going "the black quote was a mistake", the Freakonomics crew has given Myhrvold a forum to defend the quote!?!

Now there are many MANY problems with solar replacing coal, from the energy storage issue to potential use of rare-metal components. Solar is not a panacea, and the real conclusion that one reaches is nuclear power, and lots of it. I can easily construct a solid argument that solar is not viable for most of our electricity needs:

  1. It doesn't work at night/cloudy conditions without additional energy storage.

  2. It is vastly more $/W to manufacturer than a coal plant, when you include the cost of energy storage.

  3. Many solar technologies (eg, Nanosolar's thin films) involve very rare metals (eg, indium). It is unclear what a real ramp-up of solar production would do to that market.

and that QED: we need nuclear power. (It is fun to taunt greenpeace with this, BTW).

But instead Myhrvold defends his position badly...

He repeats the black canard, without mentioning that 1kWh of coal-energy also releases 1kWh of thermal energy, so unless you are placing the solar panel on a surface with albedo less than .3, even just the thermal heating argument is false. Or if you use an alternate approach ("White roof is 1T C02 per 10 m^2 annual savings equivelent" Akbari's estimate), you are still talking the CO2 load of just 500 kWh/yr of a coal plant. If your 10 m^2 roof generates 1.5 kW for 6 hours/day, that is $5/W to <$2/W. Since so much of the cost of the cells for the study is the refining of silicon, there is probably a similar drop in kWh of construction per watt of power. He misses one of the huge reasons why the efficiency crowd want buildings to have a high albeido: simply to lower the AC bill for free, and thus why you should put solar cells on the roof of your garage rather than the house itself...

He compares the cost of running a coal plant with the cost of building a solar plant, neglecting that we need to construct vastly more power plants to both meet growing demand and to deal with end-of-life on old, inefficient plants. Even then, the breakeven point is less than 3 years, by his inflate-the-cost of solar figure!

What I don't get is why they are taking this approach.

It would be easy enough to go "Whoops, the 'because they are black' quote was taken a bit out of context as a joke, thats really minor all things considered. The real reasons solar is not a panacea relate to energy storage, etc..., its being corrected in the second printing." The conclusion thus stands, but the argument becomes sound. So why defend it stupid? Is it simply trolling for attention?

Instead, what is happening is I have to conclude that anything Myhrvold says has to be assumed to be false until proven otherwise, and by unquestioningly accepting his assumptions, anything Drubner and Levitt say may need to be taken the same way.