High-income Households Pay a Large Share of US Taxes—But This Doesn’t Make Our Tax System Progressive

Perhaps the highest priority issue on the conservative agenda is keeping taxes on the highest-income Americans low. Their arguments essentially boil down to claims that raising taxes on “job creators” would doom the economy, and, besides, the government has taken so much from them already that there’s no more left to take.

In the past few weeks, the claim has been making the rounds that the U.S. tax code is already overly progressive, and much more redistributive than those of other developed nations. The line of argument following from this often seems like a tacit—or not so tacit—warning to progressives: If you want to increase public spending you’ll have to raise taxes on the middle class—because the rich just don’t have any more to give—and you’ll get hammered politically for doing so. Therefore, we can’t have any new spending. QED.

The idea that the U.S. tax system is already too progressive pops up every few years, but its persistence doesn’t mean its intellectual underpinnings are reliable. Indeed, the most recent resurgence of this theory is built on a deeply flawed foundation, consisting of a truly terrible measure of “progressivity”—namely, the fact that a relatively large percentage of all U.S. taxes collected is paid by the highest-income filers, relative to our advanced-economy peers.

As Veronique de Rugy put it three years ago: “The richest 10 percent of U.S. households (those making $112,124 or more) contribute a greater share of taxes (45.1 percent of all income taxes) than their counterparts in any other industrialized nation.” Earlier this month, Cathie Jo Martin and Alexander Hertel-Fernandez applied the same flawed metric to claim that our tax code is very progressive from an international perspective, writing that “household tax progressivity measures how much more (or less) of the tax burden falls on the wealthiest households, compared to households at the middle and the bottom.”

This flawed notion that the wealthy’s current contribution toward the sum of all taxes paid is large and shouldn’t necessarily increase was again implied by former Joint Committee of Taxation Chief of Staff Ed Kleinbard’s in a widely-circulated New York Times op-ed earlier this month. While presenting a generally sound argument about how the spending side of the federal government’s ledger should be more progressive, Kleinbard manages to grab onto the not sound argument that we must not “soak the rich” as “the American tax system already is the most progressive in the developed world.” He goes on to say that even though new government funding is needed, a “disproportionate share of those new revenues” should not be collected “from the highest income taxpayers.”

However, it is abundantly clear that the best place to get new revenue is from those that possess the largest share of it.

And in the United States, the wealthy truly do have an outsized share. The problem with thinking of “progressivity” as simply the share of total taxes paid by high-income households should be obvious: This high share is driven in large part by the fact that, relative to our advanced-country peers, the United States has more pre-tax-and-transfer inequality, not because the United States has an onerously redistributive system or because the well-off pay exceedingly high rates. The top 10 percent of the income distribution pay a larger portion of taxes in the United States because they earn a much-larger share of total income compared to their peers in other countries. The latest data, from the 2013 tax year, show that the top decile pays fully 53.3 percent of all federal taxes (including individual income, estate, payroll, and corporate taxes). That certainly seems like a lot—until you realize that the same group earned 48.2 percent of all income in 2012, according to the Piketty-Saez database. As Mike Konczal points out, even a flat tax (i.e., one that raised 48.2 percent of tax revenue from the top decile) would be deemed extremely progressive using Martin and Hertel-Fernandez’s definition. Since the top 10 percent receive a vastly disproportionate share of income, it is hardly shocking that they pay a disproportionate share of total taxes. This really doesn’t tell us much about progressivity.

Looking at just the top 1 percent is even more instructive. According to the Congressional Budget Office, between 1979 and 2010 this group’s share of total federal taxes paid (including income, social insurance, corporate, and excise taxes) increased from 14.2 percent to 24.2 percent, even as the total average federal tax rate of for the top 1 percent fell from 35.1 percent to 29.4 percent. Again, just because the wealthy now contribute a larger share of all taxes paid than in the past, it doesn’t necessarily follow that our tax code has been asking increasingly more from the richest households and thus cannot be made any more progressive.

We should grant a useful point made by Martin and Hertel-Fernandez and Kleinbard: lots of the redistribution undertaken by the public sector happens on the spending side, not the tax side, and progressive spending can and should be increased. This highlights that revenue adequacy should be an important goal. But this revenue has to come from somewhere, and in the United States there is plenty of room to raise a good chunk of it from the highest-income households who have seen the lion’s share of income gains over the past generation. We shouldn’t let a flawed measure of progressivity obscure that.


Clarification: In his New York Times op-ed, Ed Kleinbard does not employ the specific progressivity metric derided here as fatally flawed—namely, the contribution of high-income households toward total federal taxes raised. So upon reflection, it’s not totally fair to lump his argument in with these others. I do still disagree with Kleinbard’s reticence to raise more revenue progressively, but that’s a larger debate to be had later.