After running all the numbers under conditions that were very, very favorable to Mitt Romney's tax plan, the nonpartisan Tax Policy Center concluded that there was simply no mathematical way for Romney to fulfill all his promises simultaneously.
But more recently, Harvard economist Martin Feldstein has said they're wrong. Feldstein leans to the right -- he chaired President Ronald Reagan's Council of Economic Advisers -- but he's a very respected, very honest economist. No one should dismiss his work.
So, how did he get to such a different conclusion? The short answer is he made some very different assumptions than did the Tax Policy Center. The long answer is that his conclusion may not be as different as you think. After corresponding with Feldstein, the TPC, and some outside economists, I think I have a handle on where the major differences are. Let's go through it.
1) The Tax Policy Center tested whether Romney could make his promises add up without cutting taxes on people making more than $200,000 a year. Feldstein lowers that annual salary to $100,000, which roughly doubles the number of taxpayers whose deductions are getting cut. His results imply a very large tax increase for people making between $100,000 and $200,000 a year.
2) The Tax Policy Center assumes that folks making over $200,000 a year can still use the standard deduction. Feldstein takes it away from everyone making more than $100,000.
3) In his original article, Feldstein assumed that itemized deductions would be worth 30 cents for every dollar. But since Romney intends to bring the top tax rate below 30 cents, that's not reasonable. In a follow-up, he lowered itemized deductions to 25 percent, which is more plausible, though most tax experts I've spoken to think something in the range of 23 percent is more likely. The lower this number goes, the less money you can raise by closing deductions and loopholes.
4) Feldstein assumes fairly large, and very positive, growth and behavioral effects from the tax cuts. But he doesn't assume negative effects. Most models -- including, as I understand it, TPC's -- assume that as you cut deductions, taxpayers who were managing their finances to take advantage of those deductions stop doing that. That makes the deductions effectively worth less money, and makes it harder to pay for tax cuts.
5) The Tax Policy Center was using projections for 2015. Feldstein uses receipts from 2009. Ordinarily, that might not matter much, but 2009 was the worst year of the recession, and so it depressed incomes and made it cheaper to reduce rates. It's unclear how much of an effect this has on Feldstein's calculations, but a number of people mentioned it to me.
For all that, the two estimates don't really contradict each other. If you look at the original TPC study, they say that if you eliminated all deductions (except for those benefiting savings and investment, which Romney has pledged to keep) starting from the richest taxpayers and going down the income scale, you could get the numbers to work when you got to taxpayers making around $75,000.
By using somewhat more extreme assumptions in terms of growth and eliminating the standard deduction, Feldstein has gotten that up to taxpayers making round $100,000. That's a matter of degree more than kind.
So here's where we are: It might be mathematically possible to make Romney's plan work by sharply increasing taxes on people making between $100,000 and $200,000 so you could cut them on very rich taxpayers. It's not possible to make the numbers add up if you refuse to raise taxes on people making less than $200,000.
It's also worth remembering that "mathematically possible" is not nearly the same thing as "politically realistic" or "good idea." Feldstein, for instance, assumes the elimination of the charitable deduction for these taxpayers. That's something the Republican National Platform expressly promises Republicans will never do. He assumes the elimination of the standard deduction for these taxpayers, which is, again, very difficult to imagine. Even something as simple as wiping out the home mortgage-interest deduction for wealthier taxpayers seems unlikely for an administration hoping to see a housing recovery.
In my view, TPC's assumptions are closer to those of others in the political system, and so I'm sticking with "mathematically impossible" as the correct descriptor for Romney's promise. That said, there would be an easy way for Romney to end this debate: He could come out with a plan saying exactly what he'll do, or even just a set of principles -- like no tax increases on anyone making less than $100,000 -- that gives us a hint of how he makes the numbers work.