Carbon taxes won't work. Here's what will.

Or, "In which I am regrettably forced to heartlessly massacre the entire Pigou Club, of which my advisor has just declared himself a member." (Actually, this was the original title of the post, but I realized it wouldn't fit in a Mark Thoma link bullet, so...)

It fills me with great sadness to write this post, since I think that global warming is a huge threat, that economists should think more about externalities, that Pigovian taxes in general are a great idea, and that most of the people in the Pigou Club are intelligent, well-meaning, insightful, civic-minded folk, i.e. exactly the type of folks we need more of in this country. But I can no longer in good conscience stop myself from making the following argument:

Carbon taxes are not going to work.

Instead of one simple overwhelming reason for thinking this, I have a number of reasons that all pile themselves into a mountain of evidence. In no particular order, here they are.

1. Carbon taxes are politically infeasible in the U.S. A few people have tried to introduce carbon tax bills. There has been essentially no interest. This may be because there is no concentrated special-interest constituency for carbon taxes (the Pigou Club notwithstanding), or because politicians instinctively realize the existence of some of the other reasons I'm going to cite. Also note that the more well-known, Obama-supported "cap-and-trade" idea also went nowhere fast.

I cited this reason first because it's the weakest; political infeasibility should not stop economists and other intellectuals from recommending the right policy.

2. Global carbon taxes present a basically insoluble coordination problem. The U.S. emits about 1/6 of global carbon emissions. Cut U.S. emissions by 30% (a huge cut) and you cut global emissions by 5%; not enough to make a dent in global warming. Now realize that China, which now emits about 1/4 of global carbon emissions, has now twice scuttled international efforts to coordinate a reduction in carbon emissions (first at Copenhagen and then at Durban). How about Europe and Japan and Russia and Canada? Well, they agreed to carbon restrictions at Kyoto, and then promptly broke their quotas and went right on increasing their emissions.

Coordination problems are really really hard.

3. Carbon taxes are undermined by free trade. If you put a tax on carbon-emitting activity in the U.S., it'll raise the domestic price of (for example) coal. This will provide an incentive for U.S. coal miners to export their coal to other countries, especially China, as they are now trying to do. It will also provide an incentive for Americans to buy more imports from countries where it is still cheap to burn coal (e.g. China). In other words, if you tax the burning of American coal by American companies, you will increase the burning of American coal by Chinese companies, and the de facto burning of Chinese coal by American consumers. These effects will not completely cancel out the effect of a U.S. carbon tax, but they will work against it substantially. The only way to stop this would be to tax both carbon exports and the implied carbon content of imports. This would lead to big rises in tariffs.

It is hard to imagine the Pigou Club, most of whose members support free trade, uniting around the import/export tariffs needed to make a U.S. carbon tax work.

(Addendum: And you know what the U.S. can't tax? Cheap-carbon Chinese-made products replacing expensive-carbon U.S.-made products in global markets.)

4. The economic costs of carbon taxes are very hard to estimate. Pigovian taxes work by balancing the benefit of the tax with the costs. The cost of a carbon tax is the economic activity that is curtailed by the tax. But this is very hard to measure. Taxing carbon would tax driving, which would make commuting much harder for a lot of the Americans who have chosen to live in suburbs. A big carbon tax could conceivably make those suburbs economically unviable, through agglomeration effects, requiring much of the country to completely overhaul its physical infrastructure. Those agglomeration effects are very hard to predict, and might be very large. And I almost never see economists thinking about agglomeration, urban, or spatial effects or taking them into account.

Now I must admit that this reason is slightly disingenuous, since I think the benefits of higher density are massively unappreciated, and that it would be a good thing for us to dismantle much of the far-flung exurbs and replace them with walkable neighborhoods accessible by public transit. But I think that once the members of the Pigou Club start seriously thinking about the infrastructure costs associated with taxing driving, they will be less sure of their case.

5. Carbon taxes can be revoked in the future. When it comes to global warming, what matters is the total amount of carbon in the atmosphere, not the rate of emissions. Carbon stays in the atmosphere for anywhere from decades to centuries, and maybe even a lot longer than that. If a carbon tax passes at a time when the harm is relatively small, but is revoked as soon as fossil fuel prices go up (due to increasing extraction costs, i.e. "peak oil" and "peak coal"), the total effect of the carbon tax will be close to zero. And I think there is ample evidence that carbon taxes would be revoked as soon as voters started feeling pain at the pump, since "cut gas taxes" is the first thing you hear politicians saying when gas prices go up.

6. A little bit of emissions reduction is barely better than none at all. As mentioned in the previous point, carbon stays in the atmosphere a long time. What this means is that to work, carbon taxes have to reduce emissions a lot; for you nerds out there, the benefit of emissions reductions is highly convex in the amount of reduction. When it comes to carbon, a little bit of reduction is essentially no better than none at all.

And I think the previous 4 reasons are sufficient to show that a U.S. carbon tax would only reduce global emissions by a little.

There you have it: no single deal-breaker, but an accumulated weight of reasons to think that a carbon tax won't have any significant impact on global warming. The Pigou Club is a good bunch and their idea is a good try, but it just won't be enough.

So what do we do? Throw our hands up in despair and hope that global warming won't be as bad as many fear? No. I think that there is something our government can do to make a huge dent in carbon emissions, in a relatively short time frame.

 The International Energy Agency reports that America's carbon dioxide emissions have fallen by 450 million tons in the past five years. That is a drop of about 7.5% - not a slowing in the growth of emissions, but an absolute decrease. It's a big, big drop. Bigger, in fact, than any other country experienced over the same time period.

What caused this huge drop? Natural gas. Because of the shale gas boom, America has been switching from coal power to gas power, and gas emits less carbon for a given amount of energy. Now, this is only a partial solution, since gas does emit carbon, and since there's not enough gas in the world for every country to make this shift. But what it shows is that when there is a low-carbon technology available that is cheaper than a high-carbon technology, an economy will rapidly switch to the low-carbon technology, and emissions will plummet quite rapidly. And this will happen without any sort of carbon tax, cap-and-trade-system, or other emissions-limiting policy.

So the way for the U.S. government to beat global warming is this: Subsidize research into low-carbon energy sources. For electricity generation, this essentially means solar power. Solar costs have been falling exponentially, and even without a method to store solar energy during the night, solar is already cheap enough (with not-very-large subsidies) to replace much of fossil fuel power during the daylight hours.

Government money for research and development will accelerate this process. This means both basic research (which the government, in particular the military, does far better than anybody) for breakthrough technologies, but even more importantly, it means subsidies for solar companies that make the minor, incremental improvements that add up over time, driving costs steadily down. These subsidies should be seen not as government acting as a venture capitalist - most of the supported companies will eventually fail - but as government subsidizing the kind of incremental research that is not best accomplished by Defense Department grants.

Research is a public good. The day that solar becomes cheaper than coal will be the day that China and India start mothballing their coal mines and throwing up solar factories. Yes, this means that the U.S. will be subsidizing the rest of the world. Yes, it means that cheap Chinese and Indian competition will put a lot of our own solar companies out of business (as it is doing already). But it is worth it. Since there exists no global government, the only hope of stopping global warming is for one national government to kill the problem all by itself. Saving the climate (and reducing energy costs in the process!) will be worth a whole lot of Solyndras.

So at this point some of you may be reading this and saying "Wait a second. Noah Smith wants us to ignore a policy (carbon taxes) that is supported both by clear, simple economic theory and by some of the most eminent economists in the world from both sides of the political spectrum, and instead throw heaps of taxpayers' money at a bunch of companies that we know are going to fail?" And I say: Yes. That is exactly what I want. That may sound crazy to you, but everyone is entitled to one or two crazy beliefs, aren't they?

Update: Miles points out that global warming is not the only reason to reduce fossil fuel use; it is also useful in reducing the price of oil that petrostates like Iran receive. That's a topic for another post, but I'm not sure spiking Iran's economy is worth the sacrifices it would take to unilaterally bring down the world oil price through a reduction in U.S. demand (besides, a petroleum tax would be more effective for this purpose than a general carbon tax). Miles does, however, say "I second [Noah] in advocating scientific research as the most effective way to address global warming." Yay!

Update 2: Via commenter Amy, here is a Daron Acemoglu working paper that basically says the same thing I'm saying in this post. Yay!

Update 3: In fact, there is a precedent for government doing exactly the kind of R&D effort that I'm advocating here. The technology? Fracking, which produced the current shale gas boom. The potential payoff for solar is ever greater than shale gas (since solar never runs out and produces zero carbon), so if there's any energy technology the government helps to improve, it should be solar.