Do Carbon Taxes Just Feed the Beast?

Let's not jump to conclusions.

Countries introducing a carbon tax to cut greenhouse-gas emissions have mostly found that the policy works, and that jobs and incomes don't suffer. This leaves two further concerns: first, that carbon taxes feed the beast of bigger government, and second, that they hit the most vulnerable people hardest.

Again, the record is encouraging: Both concerns are overblown.

When governments bring in a carbon tax, the measure almost always meets political resistance. To deflect it, they usually promise to use the revenues to cut other taxes rather than increase public spending; what's surprising is that they typically keep the promise.

Read more from this series:

Part I: Yes, Carbon Taxes Reduce Carbon Emissions

Part II: No, Carbon Taxes Don't Kill Jobs

Often, in fact, they err in the other direction, cutting other taxes by more than the carbon tax turns out to collect. In the first instance, this adds to public borrowing; in the longer term, it puts downward pressure on government spending. If anything, experience suggests that a carbon tax is more likely to shrink the state than expand it.

The U.K. is one example. Revenues from its Climate-Change levy, introduced in 2001, were earmarked to pay for reductions in payroll taxes. They fell short, and an audit in 2007 found that the net effect was an unintended tax cut. British Columbia did the same. It used carbon-tax revenues to pay for cuts in a variety of other taxes, including a one-time tax rebate paid to every adult. The cuts added up to more than the new tax raised. British Columbia was so determined that taxes wouldn't rise overall that the enabling law threatened to dock the finance minister's pay if they did.

There are exceptions. Ireland, for instance, introduced a carbon tax in 2009 and promised to make it "revenue-neutral." It wasn't, and taxes in the aggregate went up. On the whole, though, the charge that carbon taxes expand the state is unsupported by actual experience. The pattern is pretty clear: Voters need a lot of convincing that higher taxes overall won't be the result -- so governments tie themselves down.

Potentially, though, this poses a problem when it comes to the second remaining objection to carbon taxes -- that they hit poor households hardest. By definition, a revenue-neutral carbon tax can't pay for spending to cushion poor households from higher energy costs.

Some such cushioning is certainly warranted. Poor households spend a bigger-than-average share of their incomes on fuel and electricity. A tax levied per unit of energy consumed -- which is the whole point of a carbon tax -- falls more heavily on them than on more prosperous households. A study of Denmark's carbon tax, introduced in 1992, found that it was more regressive than the standard sales tax; a study of Germany's carbon tax, introduced in 1999, found that low-income households were hit twice as hard as richer families.

The answer is not to abandon the appealing principle of revenue neutrality, much less subsidize energy consumption for the poor (which undoes the incentive that the carbon tax is intended to create). The solution, rather, is to direct some of the offsetting tax cuts at the poor. In the U.S., for instance, carbon-tax proceeds could be used to cut the payroll tax, which falls disproportionately on the working poor. Or some of it could pay for lump-sum tax rebates -- the poorer you are, the bigger the increase in your post-tax income.

Where such measures have been tried -- as in Denmark, the Netherlands or Australia (until it repealed its carbon tax this year) -- they've worked.

Precisely because carbon taxes arouse such opposition, their design comes in for close scrutiny: They won't be enacted otherwise. The opposition is a nuisance for carbon-tax advocates, but the scrutiny is all to the good. A well-designed carbon tax assures voters that taxes in the aggregate won't rise as a result, and that the people likely to be worst affected will be helped. It isn't just a nice theory: In an encouraging number of cases, that's how it's worked out.

--Editors: Christopher Flavelle, Clive Crook.

To contact the editor on this story:
David Shipley at