Some light economics

An Economist article (with it's own punny title) asserts that German government support of solar power actually harmed the industry:

It used to be an axiom that solar power grew steadily cheaper as time passed...By 2004, solar panels were coming very close to generating power at the sorts of prices regular grid-connected customers pay in places where electricity is expensive, such as Japan. Enthusiasts confidently predicted that solar cells would soon supplant grimy old power plants[.]...

But governments such as Germany’s, who wanted to give solar and other forms of renewable power an extra boost, began subsidising wind turbines and solar panels in order to speed their adoption. This was supposed to have three benefits: it would reduce emissions of greenhouse gases, spawn a fast-growing and lucrative domestic industry, and help to lower the unit costs of solar panels, thanks to the bigger volumes...

But Germany’s subsidy, which takes the form of a generous tariff for solar power, has had the opposite effect. So many firms rushed to install solar panels in such profusion that the world ran short of the type of silicon used to make them. The price of silicon—and thus of solar panels—rose. Many firms began to pursue radical new panel designs, simply to reduce their silicon consumption.

As criticisms of government subsidies go, this is fairly weak sauce. After all, if solar ever does go large-scale - and it most definitely will - we're going to need those big silicon production lines, and better bottlenecks today than bottlenecks tomorrow. Once the bottlenecks clear, technology will once again move toward the most efficient.

Which is not to say subsidies always do what they're designed to do. It's perfectly conceivable that solar energy firms, faced with a choice between spending on R&D and spending on marketing and capacity expansion, will be more likely to choose the latter - dollars now over technology tomorrow - if there's a big subsidy for solar. (In fact, this was the argument I expected the article to make, and I was surprised when it trotted out the weak sauce.)

It's important to note that world-changing technologies are rarely adopted because of government subsidies. No government paid people to use the internet, buy cars, or talk on telephones. Instead, what the government did in each of those cases was capacity building - it built roads, telephone lines, and internet routers. And the government financed the basic research that allowed those technologies to improve without hurting companies' bottom lines.

Which brings us back to the most important thing that governments do for the economy: provide public goods. Yes, Virginia, there are things that benefit society but that no one will pay for on their own. It's easy to show with math, easier still to see with common sense. If we want people to adopt solar technology, invest in government-funded solar power research, in research into better electricity transmission and storage, and in power grids that better support solar power.

A key element of our technology policy has to revolve around public good provision - research and infrastructure. Except our ruling Wall Street Journal clique, in their zeal to turn "government" into a dirty word, have apparently forgotten that public goods exist.