海角大神

From Wednesday's debate (emphasis added to聽NPR transcript):

MR. ROMNEY: Well, sure. I鈥檇 like to clear up the record and go through it piece by piece. First of all,聽I don鈥檛 have a $5 trillion tax cut.聽I don鈥檛 have a tax cut of a scale that you鈥檙e talking about. My view is that we ought to provide tax relief to people in the middle class. But聽I鈥檓 not going to reduce the聽share of taxes paid聽by high- income people...

鈥�look, I鈥檓 not looking to cut massive taxes and to reduce the 鈥� the revenues going to the government. My 鈥斅�my number one principle is there鈥檒l be no tax cut that adds to the deficit.

I want to underline that 鈥� no tax cut that adds to the deficit. But聽I do want to聽reduce聽the burden being paid by聽middle-income Americans.聽And I 鈥� and to do that that also means that聽滨听cannot reduce聽the burden paid by聽high-income Americans.聽So any 鈥� any language to the contrary is simply not accurate.

First, Romney says he will have 鈥渘o tax cut that adds to the deficit.鈥澛�How to reconcile this with聽not raisingburdens on 鈥渕iddle-income鈥� Americans and聽not reducing聽burdens on 鈥渉igh-income鈥� Americans鈥揼iven the聽聽of the kind of base broadening needed to support a 20% across the board reduction in marginal income tax rates (in聽addition聽to the proposed extension of the full complement of Bush tax cuts)聽andno increase in effective tax rates on capital income?

A few possibilities I see: (i) Romney is willing to back off the 20% figure for the marginal tax rate cuts; (ii) Romney is implicitly fiddling around with his definition of 鈥渕iddle income鈥� vs. 鈥渉igh income鈥� (consistent with聽that you might be able to avoid raising burdens on middle-income households as long as 鈥渕iddle-income鈥� ends at $100,000); and/or (iii) Romney is using 鈥渄ynamic scoring鈥� assumptions that assume growth effects offset any 鈥渟tatic鈥� revenue loss.聽 Some combination of those three tradeoffs is being exploited here.

Second, Romney says he is 鈥渘ot going to reduce the聽share of taxes paid聽by high- income people.鈥澛�How to reconcile this with reducing marginal tax rates and keeping capital income tax expenditures out of the tax base?聽 Well, two cautions here, noting what Romney is literally saying:

1. If the Romney plan is actually revenue losing, then maintaining the high-income households鈥� share of a smaller overall tax burden would still imply a reduction in the progressivity of the income tax system鈥撯�漰rogressivity鈥� referring to the existing pattern of rising average tax burdens (taxes paid/income) at higher income levels.聽 A聽肠辞苍蝉迟补苍迟听share of a聽shrinking聽progressive policy means the rich person鈥檚 burden, relative to his or her income, goes down more than it does for someone with lower income.聽 The reference to 鈥渟hares of taxes paid鈥� was a favorite way of talking about the (claimed 鈥渋ncreased鈥�) progressivity of the Bush tax cuts by the Bush Administration.聽 Given that a lot of the Romney advisers are the same people who created, promoted, and managed the Bush tax cuts (way back in 2001), the use of this statistic to advertise the 鈥渇airness鈥� of the Romney plan is not at all surprising.
2. Exactly who are the 鈥渉igh-income people鈥� in this category?聽 (Go back to point (ii) above, regarding the deficit-neutrality claim.)聽 As the Tax Policy Center pointed out in聽, if we change the definition of 鈥渉igh income鈥� to above聽$100,000 instead of above $200,000 or $250,000, it鈥檚 much easier to keep the burdens of this much broader category of households constant (or higher), by paying for net tax cuts on those above $200,000, with net tax increases on those between $100,000 and $200,000.聽 You can technically call that 鈥渘ot a reduction鈥� in the tax burdens of (all) 鈥渉igh-income people鈥� (meaning the aggregate category of people with income above $100,000), but most of us wouldn鈥檛 find that a sensible way to increase the 鈥渇airness鈥� of the tax system.

So Romney was very effective in Wednesday night鈥檚 debate at making his tax plan sound, contrary to the President鈥檚 claims, both fiscally responsible and fair, but that鈥檚 because he was just able to declare it without explaining the details.聽 And the President coming back with the details of the TPC analysis didn鈥檛 work as well as it did when he first touted the analysis two months ago in his campaign speeches and TV ads.聽 (CNN鈥檚 real-time sentiment meter of their sample of Colorado undecided voters recorded that point in Obama鈥檚 remarks as his lowest point in last night鈥檚 debate, in fact.)聽 And the debate moderator certainly didn鈥檛 follow up with the questions I would have.聽聽