The argument that Mitt Romney’s numbers don’t add up can be extremely detailed — or it can be as simple as just listening to Romney talk:

I won't put in place a tax cut that adds to the deficit. That's part one. So there's no economist can say Mitt Romney's tax plan adds $5 trillion if I say I will not add to the deficit with my tax plan.

Number two, I will not reduce the share paid by high-income individuals. I know that you and your running mate keep saying that, and I know it's a popular thing to say with a lot of people, but it's just not the case. . . . I will not reduce the taxes paid by high-income Americans.

And number three, I will not, under any circumstances, raise taxes on middle-income families. I will lower taxes on middle-income families.

That’s really all there is to it. Romney claims that he’ll keep taxes the same for the rich; cut taxes for everyone else; and not add to the deficit. If you believe that, well, you’ll also believe that Ronald Reagan didn’t increase the deficit with his 1981 tax cut and George W. Bush didn’t increase the deficit with his 2001 and 2003 tax cuts. In other words, Romney might as well just say that his tax plan is pixie dust, the Elder Wand and Samantha twitching her nose.

(He doesn't explicitly say that he won't raise taxes on the rich in the above quote, but he argued that point strongly elsewhere in the debate).

Now, there’s plenty more; the plan that Romney actually has campaigned on — which is a 20 percent cut in the rates, supposedly paid for by reducing tax expenditures — doesn’t actually achieve the things he claimed it would at the debate. That’s the whole question of the Tax Policy Center study, which finds that there just aren’t enough deductions and exclusions available to make the math work. The arithmetic says that either taxes go up for middle-income filers, or Romney relents on the 20 percent number, or the deficit goes up — each of which Romney has ruled out at various times, including last night.

Romney finesses this part of it through misdirection, citing “five studies” that supposedly back up his version of the math. But it’s not five studies, and Romney has disavowed the claims in them that would make his proposal add up, such as defining “middle-income” differently. And anyway, as Jonathan Chait reminds us, the studies that Romney cites do not support what he said last night: “None of these studies back up Romney’s claim that he won’t reduce taxes on the rich. They confirm that he will reduce taxes on the rich.”

But the truth is that none of the back and forth over studies really matters. Nor does the frustration over Romney’s lack of details for his plans. It all comes down whether what Romney actually said last night is true even as an abstract principle: that taxes can remain the same for some, be reduced for everyone else and still produce the same revenues. That’s the core fantasy in Mitt Romney’s tax proposal and, for that matter, in 30 years worth of Republican tax policy. And we have 30 years of evidence that it is just a fantasy.