There is a fair amount of support these days for the idea that the government should be getting involved in industrial policy. The argument is that the government can help make the U.S. become a competitive producer of green technologies. This will have several benefits, so the theory goes, including giving us something to export, creating high paid green jobs, and helping the environment. Proponents point to the example of Germany, which has used industrial policy to help create a large solar panel industry. Or they’ll point to China, whose government has a heavy hand in creating their many green manufacturers. Let’s leave aside for now the question of whether those policies are working for those countries, the question is should we do it here?
Many will point out that we are already involved in industrial policy in the energy sector, it’s just that our industrial policy favors dirty energy. The important thing is, Matt Yglesias tweeted yesterday, “that we have bad industrial policy now and should make it explicit and improve it”. I can agree with this sentiment, but I would modify it to say this: we have bad industrial policy now, and we should prove we can fix the existing problems before expanding it’s scope and aim. Now there is a lot of overlap between these two notions, and a fair bit of semantics, but there is at least one crucial difference: I’m arguing if we don’t have good reason to believe we can fix some of the current systematic problems, then just killing bad industrial policy is greatly preferable to attempting to build good industrial policy.
The problems our current industrial energy policy is that it doesn’t have, in fact doesn’t allow, the characteristics that we want of a healthy dynamic industry. In order for industrial policy to work in the long-run it needs to be able to allow subsidies to cease for products, companies, and even industries that become outdated. This means it has to be effective at identifying these industries, and capable of cutting off life-support. This is what the creative destruction of capitalism provides, and our current policies do not.
I agree with Matt that dirty technologies like coal and oil shouldn’t be given preferential treatment over clean ones like they currently are. Nor should ethanol, our largest attempt at a green industrial energy policy that almost everyone recognizes does not pass cost benefit. Yet if we can’t wind down the inefficient, environmentally devastating subsidies to these industries, why should we believe that the government will ever be able to hit undo if they accidently pick the wrong “winner” this time around, or if today’s “winner” becomes tomorrow’s “loser”?
The reason industrial policy has this problem is that it is explicitly geared towards creating jobs. Once those jobs are created, the goal for policy-makers becomes preserving those jobs. This is the antithesis of creative destruction, and a huge impediment to progress. What happens if we build this giant “green economy” supporting millions of middle class jobs and then a cheaper and more environmentally friendly technology comes along that makes them all redundant? Will the politicians who decide our allocations of energy via mandate and subsidy allow those jobs to go away and progress to occur, or will they fight tooth and nail to preserve the inefficient status quo?
For these reasons I find the idea of an NIH for green technologies compelling. Moneys are doled out by competitive grants to researchers with proven track records and good ideas, and the emphasis is on creating technologies, not jobs. But with respect to subsidies, mandates, and other command-and-control industrial policy, I don’t see how they will overcome the failure problem.