The diagonal line running from the lower left corner of the box to the upper right corner indicates a tax concentration curve for a proportional tax system. The bottom 50 percent (20 percent) of the population would pay 50 percent (20 percent) of the taxes. The Suits Index in this case would be zero. Finally, the dashed line above the main diagonal is a tax concentration curve for a regressive tax. The bottom half of the income distribution pays more than half of the taxes. In this case the Suits Index would be negative.
I constructed Suits Indices for the incremental taxes (both positive and negative) that follow the reform. The Suits Index for the environmental taxes is -0.248 indicating that this new tax levied in isolation would be a regressive tax. We are reducing a tax, however, in a progressive fashion. If we had levied an incremental tax equal in magnitude to the tax that we are eliminating (the “decrease” column in Table 6), that tax would also have been regressive (as measured by a Suits Index of -0.207). Note, however, that the regressivity of the income tax component that we propose to eliminate is smaller than the regressivity of the new tax (as measured by Suits Indices). Thus the shift is regressive (as measured by the difference in Suits Indices). The degree of regressivity is fairly small, however, as the difference in Suits Indices is near 0.
One problem with this measure of tax shifting and distribution is that we are implementing a small tax and so should not expect large changes in the Suits Index. The Suits Index is also a single summary measure of tax redistribution and does not indicate the full flavor of the change in tax burdens following a tax reform. As a final measure of the degree of tax redistribution, I construct a variable that I call a “tax shift” measure. It is the additional aggregate taxes paid by each decile measured as a percentage of the total amount of taxes that are being replaced by the new tax. The tax shift variable measures redistribution across income deciles but ignores redistribution within income deciles. For example, the tax reform modeled in Table 6 shifts $125.4 billion from income to environmental taxes. The households in the lowest decile face a higher tax burden equal to 0.9 percent of this shifted amount ($1.13 billion in additional taxes). A distributionally neutral tax reform would be a reform for which the tax shift variable equaled zero for each decile. The tax shift variable indicates that much of the variation in taxes cancels out within deciles and that across decile shifting is on the order of 5 percent of the total new tax revenues ($6.1 billion). The sixth and ninth deciles receive the greatest reductions in decile tax burden receiving nearly 2 percent (each) of the total new taxes.