Blog Post

Putting Carbon Tax Revenues to Work: Efficiency and Distributional Issues

Mar 22, 2016 | Peter Vail, Dallas Burtraw

In our previous blog post, we gave an overview of state carbon tax proposals under consideration in Massachusetts, New York, Oregon, Rhode Island, Vermont, and Washington. The Carbon Tax Center also has been watching these states and provides another good summary, and US Climate Plan has information on state activities. Today, we investigate the proposed uses of revenue for these policies and how this may lead to important efficiency and distributional consequences.

All of the state proposals direct a substantial share of the revenue, and four states direct a majority, back to households and businesses as rebates (sometimes called “dividends”).

  • Massachusetts would distribute all the revenue back to households and businesses in proportion to each sector’s share of emissions. The rebate to households would be directed in equal shares to all legal residents, except that rural residents would receive a small additional amount to account for motor fuel use. Businesses would receive their shares in proportion to employment.
  • Oregon would devote $50 million of its revenue for implementation and enforcement of the policy, and the remainder would be distributed directly back to taxpayers or dependents residing in the state who were claimed by taxpayers who filed a personal income tax return for the immediate preceding tax year on an equal share basis.
  • Rhode Island would direct 40 percent of its revenue as rebates shared equally among legal residents, 30 percent as rebates to employers based on their number of employees, 5 percent for administrative costs, and 25 percent for other climate policy efforts.
  • Vermont would allocate 90 percent of its revenue to tax credits and rebates. Of this total, a significant portion would be devoted to reducing the sales and use taxes, and the remaining proceeds would be divided between tax credits for employees for low- and middle-income households and rebates to employers based on their number of employees. The remaining 10 percent would be allocated to supporting low-income weatherization. (Vermont has a second legislative proposal that is somewhat more stringent but has less legislative support at this time.)
  • New York would use 60 percent of the revenue for climate adaptation and mitigation projects and the remaining 40 percent would be returned as rebates to low- and medium-income households.
  • In Washington, the intended use of revenues is clear in the legislative proposal but there is uncertainty about how that translates into realized outcomes, and revenue projects are still to be refined. Sponsors of the measure, on the ballot for November 2016, estimate that 76 percent of the revenue will be used to reduce the state sales tax, 12 percent to reduce its business and operations tax on manufacturing, and about 12 percent to provide a working families rebate (providing a tax credit of up to $1,500 for 400,000 low-income households).

It is well known that low-income households spend a greater proportion of their disposable income on energy-intensive goods such as gasoline. Hence, many analysts are concerned that putting a price on carbon will be especially burdensome for low-income households. Directing revenues back to households can mitigate the direct cost of a carbon tax. An equal per capita rebate can shift the relative burden of the carbon price away from low-income households and onto higher-income households. That occurs because low-income households typically spend less in absolute terms on energy goods, so they may come out ahead if rebates are equal across the population. Moreover, this type of rebate would not change the incentives that households face to reduce emissions under a carbon tax. That is, the tax provides an incentive to reduce consumption of highly energy-intensive goods and processes, and providing a rebate preserves this intended effect.

However, reducing other pre-existing taxes that distort economic behavior, instead of providing a lump-sum rebate, could be more efficient. For example, an existing tax on labor reduces the incentive to work because it lowers the real wage received by workers. If the revenue from a carbon tax were used to reduce the labor tax, it would increase the incentive to work, other things being equal, and help offset the cost of the carbon tax on the economy. A similar effect would be obtained if carbon tax revenue were used to reduce the tax on capital income, potentially resulting in greater capital investment.

Several recent studies by RFF experts demonstrated these effects. One study used a dynamic overlapping-generations model to demonstrate the impacts of a national carbon tax of $30 per ton of carbon dioxide. This analysis, when combined with government data on income and expenditures, gives a rich picture of the potential effects of such a carbon tax. Another analysis utilized these results with a microsimulation model to demonstrate the distributional outcomes across US states and income groups. This research considers three potential ways to use the revenue generated by the policy: (1) reducing the corporate income tax, (2) reducing the labor income tax, and (3) returning the revenue equal lump-sum rebates to households. The authors find that the lump-sum rebate is progressive, meaning that the tax would have a smaller impact on low-income households and could potentially make some of these households better off. However, this approach is also likely to be less efficient from the standpoint of economic growth.

Although it is difficult to precisely infer how such results from modeling national policies translate to state-level policies, the results can help generally inform the likely outcomes of such state-level policies. A significant lesson is that how carbon tax revenue is used substantially affects who gains and who loses as a result of the policy—it is often more important than the stringency of the actual carbon price.

Each of the state proposals has the opportunity to raise a large amount of revenue relative to each state’s budget. Providing rebates (especially to low-income households) is an effective means to ensure equitable outcomes. It may also have the advantage of being stable and self-reinforcing because a majority of households are better off due to the revenue transfer. However, if states are able to use the revenue raised by a carbon tax policy to reduce existing distortionary taxes, it could lead to more economic growth. In previous writing, we have described the priority of equity and efficiency as two distinct world views on the use of carbon revenue.  This tradeoff is one of the important issues that the states will have to face in designing their policies.