Can a Carbon Tax Cut Emissions Without Hurting the Poor?

The Issue:

There is widespread consensus that greenhouse gas emissions from human activity are contributing to rising world temperatures, increasing the frequency of devastating weather events such as hurricanes, fires, floods and droughts. In response, over 3,500 economists have signed a statement supporting the position that “a carbon tax offers the most cost-effective lever to reduce carbon emissions at the scale and speed that is necessary.” And, between 60 and 70 percent of registered voters support a revenue neutral carbon tax with the revenues returned either through tax cuts or through a “carbon dividend” to households, according to a Yale-George Mason University poll. Despite this broad support for a carbon tax, some have raised concerns that a carbon tax could place a disproportionate burden on the poor or that it would not guarantee that emission reduction goals be met. But a growing body of research finds that both of these concerns are misplaced.

A well-designed carbon tax could enhance tax progressivity and environmental justice and contribute to hitting environmental goals for emissions reductions.

The Facts:

• A carbon tax raises the price of goods and services. By design, a carbon tax causes goods and services that involve higher levels of emissions to become relatively more expensive such that consumers will tend to buy less of them or substitute for others that involve a less carbon-intensive production. Hence, a carbon tax makes coal, oil, natural gas and electricity generated from them more expensive — as well as goods and services that are derived from them or depend heavily on them for their production.
• Is a carbon tax regressive? A tax is regressive if it is a larger share of a household’s expenditures for lower income households. An Axios article quotes John Podesta, a former Obama advisor, raising concerns about equity with a carbon tax: “A real disadvantage of just a pricing scheme is you can’t directly attack the environmental injustice problem.” This reflects a broad perception that a carbon tax is regressive. After all, so the argument goes, a carbon tax is a tax on energy and energy is a good that is a larger share of the budgets for low-income households than for higher income households. But spending is only one part of the story.
• While a carbon tax affects consumer prices – the source of the regressivity noted above – it also affects returns to labor and capital. And here, the tax is decidedly progressive as returns to capital fall more than the returns to labor. Not all the cost increases that result from imposing a carbon tax are passed on to price increases for the final goods. Some of the tax would be borne by the companies that produce oil and gas and by the workers and shareholders in those companies (see here). And what we know from recent research is that it would likely fall more heavily on owners of capital than it would on workers. For instance, a firm may make lower profits rather than pass on all the cost increase from the tax onto prices. Given the concentration of wealth at the top of the income distribution, this source-side effect adds progressivity to the carbon tax. 
• Moreover, most transfer payments in the United States — such as social security — are typically indexed against changes in the general price level; as a result, transfer recipients will be less burdened by the carbon tax’s impact on consumer prices. Since transfer payments are disproportionately received by lower-income households, indexed transfers also contribute toward greater progressivity. Together the impact of the tax on returns to capital and labor and on inflation-indexed transfer payments add considerable progressivity to a carbon tax that more than offsets any regressivity observed when simply focusing on changes in consumer prices (use-side impacts).  
• A carbon tax can be made as progressive as you want. The argument that a carbon tax is regressive is wrong for a number of reasons listed above. But the most significant issue is that it ignores the use of the revenue collected from the tax. What matters for progressivity is the distributional impact of a carbon tax reform, rather than the carbon tax in isolation. The tax reform is progressive when revenue is used to finance payroll tax and personal income tax reductions as well as carbon dividends along the lines of the Climate Leadership Council’s proposal. Even when revenue is used to finance cuts to the corporate income tax, the reform looks at worst proportional, but certainly not regressive. An article by researchers from Stanford and Resources For the Future finds that for the most likely uses of revenue from a carbon tax, the tax is proportional to progressive in all cases. And, not all the revenue needs to be returned through carbon dividends or tax reductions. Given the revenue potential of a carbon tax – a 2017 U.S. Treasury study found that a carbon tax starting at $49 a metric ton would net nearly $1.9 billion over a ten year period after accounting for reductions in other taxes because of the carbon tax – a carbon tax could return some of the revenue to households through dividends and tax cuts, set aside some for transitional relief to coal miners and other disproportionately affected workers, and still contribute to green investments.  
• Some have expressed concern that a carbon tax puts a price on emissions but doesn’t guarantee that emissions will be reduced. This argument is often put forward by environmental groups, many of which would prefer a hard cap on emissions. Consider this statement by David Doniger, senior strategic director of the Climate and Clean Energy Program at the Natural Resources Defense Council (NRDC):  “While a carbon tax can be an important part of a comprehensive program to cut our carbon footprint and to hold polluters accountable, it must be accompanied by strong limits on carbon emissions, including those under existing authority granted by the Clean Air Act and all other existing legal tools.” It is likely that simply enacting a carbon tax won’t suffice for meeting specific emissions goals, since it would likely require frequent legislative updating of the tax rate to respond to changes in economic conditions and shifts in the cost curves for emission reduction. Congress is not going to want to revisit carbon tax legislation frequently to ensure long-run targets are met. Given that political reality, it would be better to design a carbon tax that includes a design mechanism to achieve any desired emission-reduction goal. 
• How can we be sure that a carbon tax will help us achieve emission reduction targets? One way to do that is to include an Emissions Assurance Mechanism (EAM) in the carbon tax legislation (see here). The EAM sets a target emissions-reduction goal for a future year, say 15 years out. The EAM then tracks emissions over that 15-year period relative to a benchmark emissions path. If cumulative emissions exceed cumulative emissions along the benchmark path, the carbon tax rate grows at a higher annual rate of increase than the default growth rate. If cumulative emissions are well below cumulative emissions along the benchmark, the tax rate is held constant in real terms. Treasury, in conjunction with other government agencies, would update the tax rate based on emissions relative to the pathway. At the end of the 15-year period, a process for setting a target for the subsequent 15-year period is set. In effect, the EAM acts as a policy thermostat, with the tax rate growing more rapidly over time if emission reduction targets are not being met and growing more slowly if targets are being met or exceeded.  
• While an emissions assurance mechanism does not guarantee the target will be achieved, it raises the probability that the target will be hit. By allowing for a pause in escalation if cumulative emissions fall well below the target emissions pathway, the proposed program would have the necessary flexibility to respond to unexpected technological advances that cut emissions (such as the catalytic converter) or economic downturns (like the current one, during which emissions declined). In simulations of this proposal targeting a 45 percent reduction from 2005 levels by 2035, a study by RFF researchers finds that the probability of hitting the target increases from 58 percent—assuming a 5 percent (real) annual carbon tax rate—to 69 percent when the EAM is included. Moreover, it reduces the probability of very high emissions and reduces the uncertainty of emissions (measured as the standard deviation in the simulation runs). Including the higher escalator in carbon tax legislation should also help to reduce concerns about policy uncertainty that could undermine support for a carbon tax. Similarly, by allowing for a pause in escalation if cumulative emissions fall well below the target emissions pathway, the proposed program would have the necessary flexibility to respond to unexpected technological or economic shocks. The Climate Leadership Council has included an EAM as one of the key design elements in its carbon dividends plan. So do a number of other carbon tax bills in one form or another that have been filed in the U.S. Congress.