Our blog posts here are always followed by a list of “Possibly related posts: (automatically generated).” In my early days at TIME these were embarrassingly useless, but now they’re pretty good, so when I saw the link under my post on Judd Gregg’s dubious tax math to something headlined “Taxes – So, is the U.S. tax system regressive or progressive?” I clicked on it. The link was broken, but a little googling sent me to the actual post, by Georgetown business school professor Ken Homa. It’s a fair-minded look at the federal tax system that concludes:
The bottom line: all of the components are progressive: federal income taxes, estate taxes, payroll taxes. So, it logically follows that the combined program is progressive.
Progressive in this context bears no ideological freight. It simply means that if your income is higher, you pay a higher percentage of it in taxes. The only controversial point here for anybody who has looked at the data is on payroll taxes. Strictly speaking they’re not progressive: because Social Security only taxes the first $106,800 in annual earnings, those in the top income quintile pay a smaller percentage of their income in payroll taxes than those in the bottom four quintiles. Homa’s argument is that because Social Security benefits are doled out according to a pretty progressive formula, the program works out to being progressive overall. I’m willing to buy that, up to a point. Somewhere in the top 10% or 5% of the income distribution things must turn regressive, as they do in the top 1% for income taxes and overall federal taxes. Still, on the whole he’s undeniably right: the federal tax system is progressive, albeit with a modest regressive kink at the very top of the income distribution.
Anyway, all of this is a longwinded prelude to a Robert Reich blog post that Homa links to, in which Reich claims that:
Viewed as a whole, the current tax system is quite regressive.
Reich offers no numbers to back this up, just the assertion that payroll taxes and state and local sales taxes are regressive. Which is true (sales taxes are regressive because high earners tend to spend a smaller portion of their incomes). But on the federal level, payroll tax regressivity isn’t nearly enough to counteract the progressivity of the income tax and the estate tax, a fact that is frequently and extensively documented by the Congressional Budget Office. There is no such handy up-to-date summary of the distributional impact of state and local taxes. The best thing I could find was a 1994 study published by the right-leaning but reality-based Tax Foundation, which found state taxes on the whole to be moderately progressive. I’m willing to believe that in states that rely heavily on sales taxes (Tennessee is the champ, with an average state and local sales tax rate of 9.36%, according to the Tax Foundation), this might not be the case. But viewed as a whole, it appears to be nonsense to say that our tax system is “quite regressive.” It’s not as progressive as Robert Reich would like it to be. It may not be as progressive as it should be. But it’s not regressive.
I’m not trying to say this is equivalent to the Wall Street Journal editorial page’s decades of misinformation on taxes (a tradition begun by the late, lamented Jude Wanniski—although in the beginning I think Jude was actually trying to tell the truth). It is misinformation, though, and when I find easy-to-refute misinformation I like to try and squash it.
Update Yay, Lane Kenworthy (see comments)! He points to a more recent Tax Foundation study on the progressivity/regressivity of state and local taxes (I searched for a while on their site yesterday and didn’t find it, then called their PR office but didn’t hear back), and has written his own blog post on the topic. Here’s what the Tax Foundation says effective state and local tax rates were, as of 2004, across income quintiles (this does not factor in the distributional effects of government spending):
Bottom quintile, 7.9%; second quintile, 10.3%; third quintile, 10.9%; fourth quintile, 11.2%; top quintile, 10.3%
When they added in federal taxes, here’s what they got:
Bottom quintile, 13%; second quintile, 23.2%; third quintile, 28.2%; fourth quintile, 31.3%; top quintile 34.5%
So state and local taxes are modestly progressive up through the bottom four-fifths of the income distribution, then turn regressive in the top fifth (and I bet that regressivity starts looking pretty dramatic when you get into the top 10%, 5%, 1%). Meanwhile, the tax system overall remains progressive, but most of that progressivity is found toward the bottom of the income distribution. Once you get to the top three-fifths it’s really pretty flat. And obviously, given that federal taxes turn regressive within the top 1% of the income distribution, I imagine that overall taxes must do that somewhere in the top 5% or 10%.
The verdict: “Quite regressive” is still an inaccurate description of the American tax system. “Not all that progressive, especially when you’re talking about the middle class and up,” is correct. So is “regressive among those making six figures and more.” So is “easy on the very rich.”