With the United States facing the expiration of a slew of tax cuts in 2013—the dread "fiscal cliff"—there has been plenty of interest in offbeat tax-reform proposals. And one idea that a few economists keep knocking around is a fee on carbon emissions. After all, if we need to raise revenue, why not just tax global-warming pollution?

Smokestacks in 1942, back before anyone worried about carbon pollution--or fiscal cliffs. (Alfred T. Palmer/Library of Congress)

A new paper from the MIT Global Change Institute lays out how a carbon tax might work in practice. The authors model what would happen if, this December, Congress enacted a small fee on carbon emissions to fend off a portion of the tax hikes and spending cuts that are scheduled to occur. The carbon tax would be levied directly on fossil fuels—on coal that comes out of the mine, say, or oil that's shipped in from overseas—and would start at $20 per ton of carbon in 2013, rising 4 percent each year thereafter.

The authors, Sebastian Rausch and John M. Reilly, estimate that this tax would raise $1.5 trillion over the next 10 years. If that revenue were then used either to cut income taxes, reduce payroll taxes, or deflect cuts to social-spending programs, the MIT authors find, most Americans would be slightly better off than if Congress simply let the fiscal cliff hit, with the Bush tax cuts and payroll tax cuts expiring automatically. (Using the carbon tax in this way would lead to an 0.02 percent bump in consumption and leisure over time.)

Now, the betting line is that Congress will do something to avert the looming tax hikes and spending cuts, so this isn't a terribly realistic scenario. Implementing a carbon tax next year would still hurt the economy—it would just hurt slightly less than the fiscal cliff. Still, the broader point of the MIT analysis is to suggest that a carbon tax could be a more economically beneficial way of raising revenue than, say, payroll or income taxes. So even if Congress waited until 2014 or 2015 or whenever the U.S. economy has recovered, replacing other taxes with a carbon tax could still provide a minor economic boost (see the first graph below).

A carbon fee usually gets criticized for hurting poorer Americans the most—they spend the biggest slice of their income on gasoline and other energy-intensive products, after all. But Rausch and Reilly found that a lot of the distributional effects depend on what Congress does with the revenue, as shown in the chart below:

CTCorp = carbon tax used to cut corporate tax rate, CTPersInc = carbon tax used to cut income tax rate, CTPayroll = carbon tax used to cut payroll taxes, CTTransfer = carbon tax used to bolster social welfare programs like Medicaid

The green line shows how different income groups would be affected in 2015 if the carbon tax was used to fend off cuts to social welfare programs like Medicaid. Lower-income Americans would benefit significantly, while wealthier Americans would take a small hit. By contrast, the red and blue lines show the effects if revenue from the carbon tax was used to cut the corporate tax or personal income tax—in those cases, higher-income Americans would come out ahead. If, however, a carbon tax was used to cut payroll taxes—that's the black line—then the welfare effects in 2015 are more or less neutral.

The MIT report argues that a carbon tax would accomplish a few other things as well. Fossil-fuel use would go down, oil imports would shrink slightly, and U.S. carbon-dioxide emissions would decline. On that last point, however, it's worth noting that the carbon tax proposed by the MIT study only gets the United States a fraction of the way toward its long-term climate targets, as shown in this graph:

Blue line: MIT reference case with no carbon tax. Black line: EIA reference. Green line: Scenario with MIT carbon tax in place.

With the carbon tax proposed by MIT (that's the green line in the chart above), U.S. emissions would be 14 percent below 2006 levels by 2020 and 20 percent below 2006 levels by 2050. That's lower than if there was no carbon tax at all.

Yet the United States would still fall short of its long-term climate goals, which involve an 80 percent cut in emissions below 1990 levels by mid-century. According to MIT calculations, a modest carbon tax, on its own, wouldn't get the United States close to that longer-term mark. It might make sense as a more economically efficient way of raising revenue. But the tax would either have to be hiked dramatically or combined with other clean-energy measures in order to make a significant dent in tackling global warming.