A new book by economists Emmanuel Saez and Gabriel Zucman is receiving a full red-carpet rollout by left-leaning media. “The Triumph of Injustice,” which claims to show that the wealthiest Americans pay lower effective tax rates than the poorest Americans, has spawned not one, not two, but three columns in The New York Times and at least three more in The Washington Post in the week leading up to its release. In reading these columns, one would have little inkling that the Saez and Zucman data simply doesn’t match other analyses.
In fact, someone reading these columns would be inclined to think that no one had ever bothered to check on the progressivity of the American tax code before. But that’s not the case—it’s just that the data isn’t usually particularly interesting to most people.
The Congressional Budget Office (CBO), for example, analyzes this very thing each year. The most recent CBO data shows that the wealthiest 1 percent pay an average effective rate of 33.3 percent in federal taxes, while the bottom 20 percent pay just 1.7 percent. That data receives few opinion columns dedicated to it because it’s about what most people would expect.
Saez and Zucman do include state- and local-level taxes in their analysis, which are, generally, more regressive than federal taxes—however, not nearly enough to make up the kind of gap that exists between their data and the CBO’s. Jason Furman, the chair of the Council of Economic Advisors under the Obama administration, analyzed federal taxes combined with state and local taxes. He found that, overall, the top 1 percent pay an average effective rate of around 40 percent, while the bottom 20 percent pay just over a 10 percent tax rate in total.
The reasons for these differences are fairly clear to see. One major issue with Saez and Zucman’s data is that they assume that corporate taxes affect only shareholders. As a professor at Columbia University has noted, this assumption differs from every other tax scorer, including the CBO and the Joint Committee on Taxation, and substantially skews the data.
Another reason for the vastly different results is that Saez and Zucman decline to factor in refundable tax credits such as the Child Tax Credit and the Earned Income Tax Credit in their analysis, instead considering them to be “transfers.” Traditional analyses do factor in refundable credits, in large part because refundable credits were introduced in large part to offset the regressivity of other elements of the tax code, such as payroll taxes. Excluding refundable credits to try to prove that the tax code is regressive is a bait-and-switch.
There are other serious problems with Saez and Zucman’s methodology as well, problems that aren’t present in other analyses in the field. Part of the reason that Saez and Zucman estimate such low tax rates for the top 0.01 percent (just 23 percent, according to them) is that their estimate of the total income of that group is likely wrong. The Internal Revenue Service data on this group is several years out of date (membership in which, by the way, is constantly changing), leaving Saez and Zucman to rely on sources like Forbes, which can be unreliable for calculations.
This unreliable denominator is part of how Saez and Zucman’s estimates differ so greatly from others. David Splinter, an economist at the Joint Committee on Taxation, estimated that this same income group pays a tax rate closer to 50 percent.
Simply put, for every questionable assumption that Saez and Zucman make, the result moves into the direction of their desired policy outcome: higher taxes on the rich.
Yet despite these massive, glaring problems, left-wing outlets continue to publish puff pieces talking these questionable numbers up, with little mention of methodological issues. All three New York Times articles make no mention of these questions. David Leonhardt, whose opinion column first broke this data, called Saez and Zucman’s book “the most important book on government policy I’ve read in a long time.” PolitiFact even rated a claim by Rep. Ilhan Omar (D-MN) that cited this data “half-true,” but only because she mixed up the top 1 percent and the top 0.01 percent.
Saez and Zucman’s data will undoubtedly drive a great deal of tax policy discussion on the left for the foreseeable future. But it’s important to note the data set behind this discussion is deeply flawed, or at the very least controversial—even if left-leaning outlets won’t.