I seem to have struck a nerve with my criticism of the recent Tax Foundation report on the redistributive effects of U.S. government tax and spending policy.
Since I posted my criticism, the Foundation rushed out a new appendix section to the report and a FAQ, and Andrew from the Tax Foundation left a set of comments on The Picket Line to try to explain away the problems I found.
These new explanations, though, don’t much impress me. There’s a lot about how the approach they chose to use with the data has a long and respectable history in decades of public finance scholarship, which may very well be true for all I know, but I was hoping for more than an appeal to authority in response to what I thought were pretty basic substantive criticisms.
And there’s a lot of dishonest straw-manning of my arguments, which is also disappointing.
What it comes down to is that the best you can say for the report is that the Tax Foundation may have come up with some intriguing numbers based on some imperfect methodologies in an area of fiscal studies where imperfect methodologies are about all we have. However, they lose my respect by trumpeting their study with language and graphs that distort this and that make claims for the study that go way beyond any it has a right to assert:
Indeed, as the report would have it, we are a nation of a minority of taxpayers and a majority of “tax consumers”! This is just simply a dishonest description of the data they present. They do not show that 60% of Americans “receive” more from the government than they give it, nor that they are the net beneficiaries of “redistribute[d]” “income”.
This sort of description is the result of making all government spending the one-to-one dollar benefit equivalent to somebody of money taken from them in taxes. Thus, a person’s “net tax burden” becomes the difference between what they pay and what the government spends on their behalf (including all government spending, not just benefits paid to that person).
To see how this methodology can lead to crazy results, note that if the government doubled the amount of taxes it collected from each of us and then spent all this new money to have the face of George Washington engraved with lasers on the surface of the moon, by the standards of this Tax Foundation report, the result of this would be that 60% of America would see their “net tax burden” fall even though each one of them saw their actual tax burden double!
In fact, the government could take all of everybody’s money, spend it on some crazy thing like this, and the Tax Foundation would consider it a redistribution of wealth from the rich to the poor by which 60% of Americans would be judged to be “tax consumers” benefiting from income redirected from the taxpayers in the upper 40%.
Because taxes and spending exactly cancel out, no matter how much the government takes in in taxes, the sum total of the “net tax burden” over the country as a whole is always zero (or would be if there were a balanced budget). That is a remarkable discovery! In fact, since budget deficits have been popular lately, there is more government spending than tax revenue, and the nation’s “net tax burden” as a whole is less than zero! Grover Norquist, you may retire!
The Tax Foundation isn’t humble about such limitations of its methodology, but rather it boldly suggests that its weird discoveries about “net tax burden” have important public policy ramifications when, for instance, assessing the progressivity of the tax code: “Any amount of progressivity can be achieved by some mix of tax and spending changes.” In other words, if for some reason you have a hankering to make the tax code less progressive, you can make up for this by spending more money on “public goods” since this spending is inherently progressive! If you spend more money on the military, it’s exactly as though you were taking money from the rich and giving it to the poor, so don’t feel bad if you raise the tax rates on the poor to compensate!
And this was all done in the most respectable traditions and practices documented in decades of fiscal incidence literature, so you don’t just have to take their word for it.
