Ezra Klein directs us to a new Tax Policy Center analysis of the Simpson-Bowles — by the way, have you noticed how some people call it Simpson-Bowles and others call it Bowles-Simpson? My instinct is to go alphabetical, but that’s just me — tax reform plan:
The Tax Policy Center just released the first distributional analysis on the commission’s first tax option: zeroing out the tax code and restoring only the child-tax credit and the Earned Income Tax Credit — both of which are thought to be the most progressive of the bunch. The results aren’t good for the commission. Under this option, the poor pay more and the rich pay less.
I’m a great admirer of the Tax Policy Center, but I’m afraid there might be some confusion here about how to think about the distributional impact. Consider that high-earning households are more likely to have children. The bottom quintile households have a disproportionate number of single-person households. And so it’s not too surprising that the child tax credit wouldn’t benefit households evenly, or rather that it would prove somewhat more beneficial to households in the top quintile than the bottom four.
Low-income married couples with children have seen a marked decline in their taxes since 1970 (see figure 2). For example, the average combined income and payroll tax rate for married couples with two children and income at the poverty level fell from about 7 percent in 1970 to negative 17 percent in 2006. That decline resulted in large part from the creation and subsequent expansion of the refundable EITC and partially refundable CTC.
Because the earned income tax credit (EITC) is refundable and the child tax credit (CTC) is partly so, the average effective individual income tax rate for the bottom two income quintiles in 2007 was negative; that is, the tax credits more than offset positive income tax liability, so that the average household in these quintiles received a net payment from the government.
It’s possible that tax levels for the bottom two quintiles can’t realistically go much lower. This doesn’t mean it’s totally awesome to see the tax take go up in both quintiles. But I’d suggest that a system that generates more revenue more effectively by slightly increasing the tax take in the bottom two quintiles might be part of a broader tax-and-transfer system that is more progressive overall. That is, looking at the impact of the tax system in isolation could prove highly misleading, given how much of the income of the bottom two quintiles comes from transfers, as the TPC explains:
Low-income households face a lower than average effective payroll tax rate because they get less of their income from earnings and more from transfer payments than do higher-income households. In 2007, payroll taxes claimed 7.3 percent of the cash income of tax units in the lowest quintile, compared with 8.3 percent for all tax units.
This certainly isn’t dispositive. But I’d say it’s valuable information.