The Tax Policy Center (TPC) has been banged up pretty badly for its recent tax study asserting it is “impossible” for Mitt Romney to reduce rates across the board by 20 percent while keeping it revenue neutral and not raising taxes for those earning less than $200,000.

Right Turn first exposed the sleight of hand Aug. 7. I wrote then that “if you assume a bunch of stuff, include a bunch of stuff that Romney never said and then leave out all of Romney’s other policies, then Romney’s plan doesn’t work. TPC does this in the guise of ‘assumptions,’ which President Obama’s ads and pro-Obama pundits leave out.” Other critics soon followed.

Well, TPC obviously had no choice but to concede that some “assumptions ” do lessen the amount of tax expenditures available to “pay for”the lower rates. Two of those assumptions that I (followed by other TPC critics) looked at were exclusion of muni bond interest and exclusion of interest on life insurance savings from taxable income for upper-income taxpayers. These two items alone, the American Enterprise Institute concluded, give $90 billion more, more than enough to cover the alleged $86 billion shortfall TPC says that would have to be gotten from those making less than $200,000 to keep Romney’s plan revenue neutral.

So TPC comes back and says that it will add back some of the muni bond interest but not all of it. The authors assert that the full $90 billion shouldn’t be counted because “the total tax expenditure for the exclusion of interest on state and local bonds is $63.8 billion in 2015. However, about $18.6 billion accrues to corporations, and we have already treated this portion as eliminated to pay for corporate tax reform.” They treated it that way, but nowhere does Mitt Romney say this.

In fact, one of the fundamental and incorrect assumptions of TPC is that the authors set up an arbitrary “assumption” that none of the corporate tax changes could cover rate reductions for individual tax payers. By the time TPC gets done (and includes the lower 20 percent tax rates at which municipal bonds would be taxed) it says that this item is worth only $25 billion Together with the insurance contract interest, the authors say that these items provide only $45 billion of the $86 billion. Well, it’s getting less “impossible,” isn’t it?

TPC, however, ignores a host of other problems resulting from its “assumptions.” As I pointed out this month: “TPC eliminates the death tax but keeps the step-up-in-basis provision for heirs. You have to choose one or the other. It also treats refundable credits for low-income taxpayers as a tax expenditure rather than looking to the spending side to account for checks going out to non-taxpayers. And TPC gives short shrift to dynamic scoring. (For a comparison in the treatment of dynamic scoring, take a look at a 2011 piece by Martin Feldstein.)”

Other critics of the TPC study have pointed to other problems. Matt Jensen of the American Enterprise Institute explains: “TPC’s report offers scant documentation on how the different pieces of the tax reform are modeled and how individuals in different income groups would behave. Although it may take a little work, my dream study would provide results for a wide variety of these assumptions, or, at the very least, include a fat appendix that details the assumptions they use. This is particularly important for assumptions regarding growth and individual feedback effects because different economists, based on a broad literature, have different views that imply divergent revenue and distributional outcomes. Due to its lack of documentation, TPC’s report, in a way, is only relevant for those economists who agree with their assumptions, and they don’t even know who they are because not all of the assumptions are clear. Moreover, the lack of documentation is quite misleading for non-economists who aren’t aware of these issues.” That would include most journalists.

Grover Norquist, head of Americans for Tax Reform, put all of this in layman’s terms in an e-mail to me: “Garbage in, Garbage out.” He notes that in prior tax reform efforts when the top bracket was lowered “the amount paid by higher income folks increased and the percentage of total taxes paid by higher income folks increased.”

The revisions in TPC’s assumptions give its game away: Make certain assumptions, and you can get the answer you want. Hide the model that weights those assumptions, and no one can check your work.

This is why it was a departure from academic decorum for TPC to declare a plan about which TPC was filling in the blanks to be “impossible.” That’s what you’d expect of former White House advisers and partisan Democrats, not from people who offer themselves out as impartial think-tankers.

UPDATE (10:20 a.m.): Matt Jensen of AEI, who did some of the heavy lifting in debunking TPC’s first effort, has some thoughts along the same lines as Right Turn’s analysis. Jensen writes with tongue planted in cheek: “This new paper is quite useful and helpful, and the policy community should be grateful that they wrote it. However, as blemishes are apt to stand out up close, in a few ways this document highlights my misgivings regarding their initial study.” Read the whole thing.