Carbon taxes, to be effective, will raise large sums of revenue. But generating revenue for the government isn’t their main job. The purpose of “putting a price on carbon” is to cut carbon dioxide emissions by decisively raising prices of fossil fuels relative to conservation, efficiency and cleaner alternatives. We don’t need more tax revenue in order to cut CO2 emissions, we need to shift more of the total tax burden onto dirty energy, and to do so without harming low- and middle-income families..

Effects of Carbon Taxes Differ Across the Income Range

Most middle- and low-income households spend a larger percentage of their income on gasoline, other fuels and electricity than do higher-income households. The wealthiest 20% of U.S. households spend just 2.3% of their after-tax income on gasoline; the percentage for the lowest “quintile,” 9.1%, is four times as high. Clearly, imposing a gasoline tax or, by implication, a carbon tax, without tax-shifting or dividends, would have disproportionate impacts on lower-income families.

But the picture is quite different when energy expenditures are viewed in absolute dollar terms — arguably a more meaningful measure. In 2005, the top-echelon quintile spent an average of $3,182 on gasoline, or 3.6 times as much as the $882 spent by the poorest 20% of households. Put differently, when all household outlays for gasoline are apportioned among quintiles, the highest-earning quintile accounted for 32% of the total, while the lowest quintile contributed just 9%. (The middle quintile, true to its name, spent exactly 20% of the total.)

What’s true for gasoline applies to energy in general, as the Congressional Budget Office confirmed in early 2007 in a report, Trade-Offs in Allocating Allowances for CO2 Emissions. Examining the price increases that would ripple through the economy from raising the price of carbon emissions sufficiently to reduce U.S. emissions by 15%, the CBO estimated that households in the highest quintile would average an additional $1,800 a year for fuel, electricity and goods — more than three times the additional $560 that households in the lowest quintile would pay. (The middle three quintiles would pay $730, $960 and $1,240, respectively.) Although the lowest quintile would bear the greatest cost as a percentage of household income, it would pay the least in absolute terms, thus making possible a “progressive” outcome through rebates or appropriate tax-shifts.

The upward skew in carbon use over the income range comes about because higher-income households don’t just drive more, they also fly more (burning jet fuel), they tend to own bigger (and sometimes multiple) houses to heat and cool, and they buy and use more products that require electricity or industrial fuels to manufacture, deliver and use. This means that the bulk of carbon taxes will be paid, directly or indirectly, by families of above-average means. For the gasoline part of carbon taxes, we estimate that around two-thirds will be paid by above-average-income households. (Calculated by summing: the first quintile’s share of gasoline expenditures, 31.6%, the second quintile’s share, 25.0%, and half of the middle quintile’s share, 19.8%, which yielded a total of 66.6%. Data are from the Bureau of Labor Statistics’ Consumer Expenditure Survey, 2005.)

A Carbon Tax Can Be Made Progressive With Appropriate Revenue Return

The upward distribution of carbon tax burdens creates an opportunity for progressive tax-shifting: transferring a portion of the tax burden from regressive taxes such as the payroll tax (at the federal level), and the sales tax (at the state level) onto carbon emissions instead. Advocating a carbon tax shift, Al Gore has urged, “Tax what we burn, not what we earn.” This approach could have a net progressive effect, i.e., shifting taxes away from payroll and/or sales taxes and onto carbon pollution could raise, not lower, the after-tax incomes of a majority of below-median-income households. In addition to this potential progressive effect of tax shifting, a number of economists, [including Gilbert Metcalf, Alan Viard, Robert Shapiro… ] have suggested that use of revenue to reduce other taxes would improve the overall efficiency of the economy, for example, by removing burdens on work.

In August 2007, the World Resources Institute and the Brookings Institution jointly published a policy brief by Tufts University economics professor Gilbert E. Metcalf, outlining a national carbon tax paired with a reduction in the payroll tax. The brief assesses the impact of a tax of $15 per metric ton of carbon dioxide which is used to rebate the federal payroll tax on the first $3,660 of earnings per worker. This tax swap is both revenue-neutral and distributionally neutral. (Harvard professor and former Bush Administration economist Gregory Mankiw mentioned the Metcalf paper in a Sept. 2007 New York Times op-ed, discussed on our blog.)

An alternative approach that is unquestionably progressive, and appealingly straightforward, is to return the carbon tax revenues equally to all U.S. residents. This  so-called “dividend” or “green check” would be a national version of the Alaska Permanent Fund, which once a year sends every resident of that state an identical check drawn from earnings on investments made with the state’s North Slope oil royalties. (We suggest that for a federal carbon tax, “dividend” checks be provided at least quarterly to keep households ahead of the budget treadmill.) With carbon tax revenues returned through equal (per capita) dividends, the vast majority of lower- and middle-income households would get back more in the dividends than they would pay in higher prices for carbon-intensive goods induced by the tax.

The 2008 “economic stimulus checks” authorized by Congress and the Bush Administration offer a preview of the simplicity and potential rapid implementation of a “carbon tax-and-dividend” approach.  Within months, the government rebated $168 billion in tax revenue via checks or equivalent direct deposits. Individuals received up to $600 and couples up to $1,200 with additional payments of $300 per child.

A numerical illustration may help convey the progressive impact of returning carbon tax revenues equally to all U.S. residents:

• We estimate that the revenues from Year 1 of our graduated “Starter Tax” of $37 per ton of carbon emitted (equivalent to $10 per ton of CO2) would be approximately $55 billion a year. (The actual dollar amount isn’t critical for this example.)
• Assuming that this entire amount is returned equally to all 300 million U.S. residents, each annual “dividend” would be $183.
• From our earlier observation that the lowest-income quintile makes 9% of gasoline expenditures, $4.8 billion of the carbon tax payments would be made by the 60 million people comprising this quintile. [Using gasoline as a “proxy” for carbon fuels, for the reasons given earlier.]
• Averaged over those 60 million, each individual in the lowest-income group would spend an average of $80 a year in carbon taxes.
• Netting this $80 per-person carbon tax from the $183 per capita dividend, we see that the average person in the bottom quintile would gain approximately $100 in the first year of our $37/ton carbon tax. Annual installments to ramp up the tax rate would only add to the gains of this lowest-income group.

Using the same methodology, we calculate that carbon tax payments for the top quintile would average $290 per person per year. Since each person in this group would also receive a $183 dividend, we see that the average person in the top quintile would pay a little over $100 in net carbon taxes during the first year. The same result of the more affluent paying a net tax and the less affluent receiving a net gain applies to the second and fourth quintiles as well, though the impacts are of smaller magnitude. The middle group would average no net change in after-tax income.

We recognize that some individuals in the lower-income quintiles might be subject to higher than average carbon taxes because they drive old, fuel-guzzling cars, drive long distances to work, have old inefficient furnaces and/or live in poorly-insulated homes. A portion of the carbon tax revenues could be used to help reduce the energy use of their cars and homes, or such assistance could be financed through general revenues.

While this analysis represents only a first approximation, the principles are clear: either carbon “dividends” or a tax shift can ameliorate the disproportionate impact of carbon taxing, to the point that below-median households become net beneficiaries. CTC advocates such an outcome, for both pragmatic and ethical reasons. Pragmatically, because a carbon tax will require broad, sustained support and ethically, because it’s not tenable to solve the climate crisis on the backs of those who can least afford it.

How Much Revenue to Return?

De-carbonization of our economy to the extent needed will take decades, and require broad, even increasing political support. Just as important as setting an effective climate goal is designing a policy that will ensure decades of continued political will to drive down emissions. We are aware of the many competing demands for carbon revenue illustrated by the long list of those seeking free allowances under the cap/trade/offset bills. But instead of handing out carbon revenue — even to worthy causes — we advocate return of all (or nearly all) carbon revenue so that roughly 2/3 of households will enjoy ongoing financial benefits from a carbon tax — an important pillar to ensure sustained public and political support.

Carbon Taxes May Not Be Inherently Regressive When Their Effects On Both Income and Spending are Considered

In June 2010, four leading economists released a report carefully examining the impacts on various income groups of different cap-and-trade carbon pricing proposals. The following is the abstract of Distributional Implications of Alternative U.S. Greenhouse Gas Control Measures by Sebastian Rausch, Gilbert E. Metcalf, John M. Reilly, and Sergey Paltsev, published by the MIT Joint Program on the Science and Policy of Climate Change (June 2010):

We analyze the distributional and efficiency impacts of different allowance allocation schemes for a national cap and trade system using the USREP model, a new recursive dynamic computable general equilibrium model of the U.S. economy. The USREP model tracks nine different income groups and twelve different geographic regions within the United States. Recently proposed legislation include the Waxman-Markey House bill, the similar Kerry-Boxer bill in the Senate that has been replaced by a Kerry-Lieberman draft bill, and the Cantwell-Collins Senate bill that takes a different approach to revenue allocation. We consider allocation schemes motivated by these recent proposals applied to a comprehensive national cap and trade system that limits cumulative greenhouse gas emissions over the control period to 203 billion metric tons. The policy target approximates national goals identified in pending legislation. We find that the allocation schemes in all proposals are progressive over the lower half of the income distribution and proportional in the upper half of the income distribution. Scenarios based on the Cantwell-Collins allocation proposal are less progressive in early years and have lower welfare costs due to smaller redistribution to low income households and consequently lower income-induced increases in energy demand and less savings and investment. Scenarios based on the three other allocation schemes tend to overcompensate some adversely affected income groups and regions in early years but this dissipates over time as the allowance allocation effect becomes weaker.

Finally we find that carbon pricing by itself (ignoring the return of carbon revenues through allowance allocations) is proportional to modestly progressive. This striking result follows from the dominance of the sources over uses side impacts of the policy and stands in sharp contrast to previous work that has focused only on the uses side. The main reason is that lower income households derive a large fraction of income from government transfers and, reflecting the reality that these are generally indexed to inflation, we hold the transfers constant in real terms. As a result this source of income is unaffected by carbon pricing, while wage and capital income is affected.

(Emphasis added.)

That’s a surprising finding that runs counter to conventional economic thinking that consumption taxes (including carbon taxes) are almost always regressive. This new finding may blunt the distributional argument for progressive revenue return of carbon revenue. But tax shifting also offers the potential for efficiency gains by reducing distortionary taxes in other areas. As Dr Metcalf testified to the Senate Energy and Natural Resources Committee:

Any allocation mechanism should address the regressivity of carbon pricing ideally in a way that does not forego the opportunity for gains in economic efficiency through the possibility of tax rate reduction.

Further, as the example below suggests, the political and “perception” advantages of tax shifting or direct distribution of carbon revenue can’t be overlooked.

Revenue Return Drastically Affects Public Attitudes About Taxes and Fees

By shifting revenue back to customers, the NY State Public Service Commission created incentives to limit costly directory assistance calls that had been driving up rates for all. Cornell economist Robert H. Frank recounted in the NY Times (5/16/07):

The commission argued that a 10-cent charge for directory assistance calls would give consumers an incentive to look up telephone numbers on their own whenever convenient, which would free up operators and equipment for more valuable tasks. Although the commission’s proposal promised net benefits for the average telephone subscriber, it was greeted by a firestorm of protest… Commission officials then introduced a simple amendment that saved it. In addition to charging 10 cents for each directory assistance call, they proposed a 30-cent credit on each consumer’s monthly phone bill, a reduction made possible by the additional revenue from the charge and the savings from reduced volumes of directory assistance calls. Because this amendment promised to reduce the monthly bill of customers willing to use their phone books, political opposition vanished overnight.

Interestingly, Frank was using the “411” telephone controversy to suggest the advantages of  proposals to price traffic congestion, while here we are addressing CO2 emissions and the climate crisis. The common principle in all three situations is charging for services or activities that entail costs, and the political as well as moral imperative of returning revenue to protect vulnerable members of society from the impact of the higher charges and simply to build support.