By Gilbert E. Metcalf, Tufts University
Should we raise environmental taxes? This question is being asked more often in the face of widespread environmental problems, including global climate change. A textbook answer to this question would be yes, we should raise taxes to the point where the tax equals the marginal social damage from pollution. But measurement problems abound: How do we quantify marginal social damages? Distributional concerns also come into play because environmental taxes tend to be regressive: Poor people pay a higher share of their income in these taxes than rich people.
This paper addresses how one could design an environmental tax reform that reduces or even eliminates this regressive nature. For example, the paper shows that a modest tax reform in which environmental taxes equal to 10 percent of federal receipts are collected has a negligible impact on the income distribution when the funds are rebated to households through reductions in other taxes. Indeed, overall progressivity can even be increased as the tax reduction package is slightly modified.
The green tax reforms are then compared to a reform that shifts the tax base from income to consumption. In this case, it is difficult to maintain the level of progressivity that exists under the current income tax. Overall, however, distributional concerns need not stand in the way of the increased use of environmental taxes.
Looking at Environmental Tax Reforms
In recent years there has been a great deal of interest in the use of environmental tax revenues to substitute for some portion of existing taxes, and the paper begins with an examination of this literature, dating back to the work of Tullock (1967) and Terkla (1984). This early literature focused on the efficiency implications of an environmental reform and led to a debate over what has been dubbed the "double dividend hypothesis" (a concept addressed in more detail in the paper in this series by Sanstad and Wolff). Most of the debate over green tax shifts has focused on this hypothesis, but distributional considerations are clearly important, and little work has been done in this area. While some authors have challenged the perception that most environmental taxes are regressive by taking into account lifetime considerations, it is clear that these concerns still limit political support for the greater use of green taxes.
Using the 1992 Input-Output Accounts, the paper traces through changes in intermediate goods prices resulting from taxes on these goods. These taxes are allocated to households in the 1994 Consumer Expenditure Survey to measure the distributional impact of green tax reforms using both annual income and lifetime income approaches. The paper analyzes a number of different reform scenarios. First, it examines an environmental tax shift equal to 10 percent of federal revenues, in which the new taxes are taxes on carbon dioxide emissions, gasoline consumption, air pollution, and the use of virgin materials in production. These revenues are used to fund three changes in existing taxes: (1) an exemption from the payroll tax of the first $5,000 of wages, both at the personal and business level; (2) a refundable $150 tax credit for each exemption taken in the personal income tax; and (3) an across the board income tax cut of four percent. Using annual income to rank households, this scenario reduces the progressivity of the tax system very slightly, but the analysis suggests that slight changes to the tax reduction package could make the overall reform progressive.
To measure this sensitivity, the reform plan is changed to include a $200 tax credit for each exemption while reducing the proportional rate reduction to maintain revenue neutrality. With this small change, the tax reform looks progressive (as measured by a summary index of tax progressivity) regardless of the income measure used or cohort considered. In a third experiment, rather than give each worker a $5,000 exemption from payroll taxes, the exemption is tied to family size. This tax reform is also mildly progressive, except in the very lowest income group. In sum, all of the revenue-neutral green tax reforms considered in the paper can essentially be viewed as both distributionally and revenue neutral, which shows that it is indeed possible to implement environmental taxes in a progressive fashion when other taxes are reduced.
Comparisons to Other Consumption Taxes
It is instructive to contrast the distributional impact of a green tax reform with another reform under discussion: a shift from income to consumption taxation. In this light, the paper considers a shift to a broad-based retail sales tax, both as a replacement of the entire income tax with a sales tax, and a replacement of 10 percent of the income tax with a smaller sales tax. This latter reform allows a direct comparison between the environmental tax reforms discussed above and the consumption tax reform.
The tax base for the national sales tax as modeled in the paper is quite comprehensive, and the analysis shows that this reform is very regressive. Tax liabilities increase for the bottom 70 percent of the income distribution and decrease for the top 30 percent when annual income is used to measure progressivity; in fact, the lowest income group sees its average tax rate increase by 34 percentage points, while the top income group's average tax rate falls by 7 percentage points. The regressivity of the sales tax is reduced significantly when lifetime income analysis is used, but the reform is still regressive: Tax liabilities still increase for the lowest 70 percent of the income distribution and decrease for the top 30 percent. The paper then analyzes a more modest sales tax reform that replaces 10 percent of federal revenue. This tax reform is also regressive, although less so than the sales tax replacement.
This paper demonstrates that distributional concerns about the greater use of environmental taxes can be addressed through a careful menu of tax reductions. While it is true that environmental reforms could be designed that are quite regressive, the analysis indicates that distributionally neutral (or even mildly progressive) reforms are feasible. In fact, if tax reformers wish to maintain or increase the progressivity of the tax system, this analysis shows that "green" tax reforms may be more effective than consumption tax reforms in achieving that goal.
About the author
Gilbert Metcalf received his Ph.D. in Economics from Harvard University in 1988. Currently he is an Associate Professor of Economics at Tufts University. In addition to his Tufts affiliation, Professor Metcalf is a Research Associate at the National Bureau of Economic Research and has taught at Princeton University and the John F. Kennedy School of Government at Harvard University. He has served as a consultant to the U.S. Department of the Treasury and the Argonne National Laboratory. Professor Metcalf's primary research area is applied public finance with particular interests in taxation and investment, tax incidence, and energy and environmental economics. He has published papers in numerous academic journals and has contributed chapters to several books on tax policy.
Order "A Distributional Analysis of an Environmental Tax Shift" ($13):