The Economist on Carbon Taxes vs Cap-and-Trade

The Economist, surprisingly, and disappointingly, has come out in favor of carbon taxes over a cap-and-trade regime. (Thanks to Greg Mankiw for the link.)

As is typical of Economist leaders, the argument seems reverse-engineered from the conclusion. So, as a public service, let me fill in some of the blanks that the Economist doesn't see fit to include.

Most economists agree that carbon taxes are a better way to reduce greenhouse gases than cap-and-trade schemes.

This is a stretch: I'd love to see what kind of evidence the Economist could adduce for it. While carbon taxes might make sense in a specific case such as gasoline taxation, there's an enormous number of economists working very hard on cap-and-trade systems who have come to the conclusion that they make much more sense than trying to cover all carbon emissions with a tax.

In the neat world of economic theory, carbon reduction makes sense until the marginal cost of cutting carbon emissions is equal to the marginal benefit of cutting carbon emissions.

Well, maybe. But in the real world, carbon reductin makes sense until – well, until global carbon emissions start going down rather than up, for starters. People vary on the amount that they think carbon emissions need to be reduced. But since "the marginal benefit of cutting carbon emissions" is something that no one is ever going to be able to calculate with any accuracy, there's very little point in trying to use it to develop optimal levels either for a carbon tax or for a cap-and-trade scheme. The best we can do is simply set a target for carbon emissions, and try to meet it as best we can.

Clearly, trying to reduce carbon emissions by taxing carbon is much harder than trying to reduce carbon emissions by simply regulating the amount of carbon that can be emitted.

If policymakers set a carbon tax too low, too much carbon will be emitted. But since the environmental effect of greenhouse gases builds up over time, a temporary excess will make little difference to the overall path of global warming. Before much damage is done to the environment, the carbon tax can be raised.

Which will raise carbon prices – and which will also mean an end to the much-vaunted predictability of carbon taxes. But it won't, with any certainty, reduce carbon emissions.

Misjudging the number of permits, in contrast, could send permit prices either skywards or through the floor, with immediate, and costly, economic consequences.

I certainly can't see "costly economic consequences" to falling permit prices. And if permit prices rise a lot, then that would just demonstrate the high level at which a carbon tax would need to be set in order to effectively reduce emissions. A level, by the way, which might well be politically unfeasible.

Worse, a fixed allotment of permits makes no adjustment for the business cycle (firms produce and pollute less during a recession).

This is exactly the wrong way around. It's the carbon tax which makes no adjustment for the business cycle. If firms produce and pollute less during a recession, then they'll need fewer permits, and the price of permits will come down.

America has had tradable permits for SO2 since the mid-1990s. Their price has varied, on average, by more than 40% a year.

This is very misleading. The price of SO2 has varied a lot mainly because it's come down a lot. It turns out that reducing sulfur emissions is much cheaper than people thought it would be. That's good news.

Extreme price volatility might also deter people from investing in green technology.
Even without the volatility, some economists reckon that a cap-and-trade system produces fewer incentives than a carbon tax for climate-friendly innovation. A tax provides a clear price floor for carbon and hence a minimum return for any innovation.

Price volatility does not deter investments, necessarily. Let's say that a green technology only becomes cost-effective when the price of carbon reaches $35 per ton. Then if a carbon tax is set at less than that (and any carbon tax would be set at less than that), no one will ever invest in that technology. On the other hand, if permits traded at less than that, there's still a case to be made to invest, on the grounds that the price of the permits might rise to more than $35 in the future.

Indeed, a carbon tax provides a clear price ceiling for carbon, and hence a maximum return for any innovation. That's hardly an incentive for climate-friendly innovation.

Ultimately, the argument is simple. A cap-and-trade system caps carbon emissions; a carbon tax doesn't. If you want to reduce carbon emissions, then you need to cap them. A carbon tax, on the other hand, is very unlikely to ever reduce emissions to 1990 levels, let alone anything lower. Auction off most of the permits, and don't allow a "safety valve". Then you get all of the benefits of a carbon tax (government revenues), and none of the uncertainty about whether or not emissions will actually, in the end, be reduced.