In the New York Times, David Leonhardt discusses what he calls “a popular talking point on cable television and talk radio”—that 47 percent of Americans no longer pay any income tax. Leonhardt grants the point—“The 47 percent figure is not wrong”—but adds, “Over the last 30 years, rates have fallen more for the wealthy, and especially the very wealthy, than for any other group.” But while the notion that something approaching half of all Americans don’t pay income taxes is true, Leonhardt’s claim about tax rates for the wealthy is not.
As I detailed in several charts yesterday, according to the most recent data released by the Congressional Budget Office (which shows figures through 2007), the top quintile of income-earners, and particularly the top 1 percent, pay a far higher percentage of their income in federal taxes than anyone else does (a point that Leonhardt grants). More to the point, over the most recent quarter-century that the CBO has measured, wealthier Americans’ federal tax rates did not fall. Indeed, they rose.
For Americans as a whole, the total effective federal tax rate (the percentage of people’s income that they really pay in federal taxes, once all of the deductions have been figured in) was the same in 2007 as it was in 1983: 20.4 percent. (Even in the wake of the 2001 and 2003 tax cuts, the percentage of Americans’ incomes that they paid as taxes was as high in 2007 as it had been a quarter-century earlier—a trend that will resume once the economy gets moving again.) But while tax rates on the whole did not change, the tax rate for the top quintile—and the top 1 percent—rose 5 percent from 1983 to 2007, while every other quintile’s tax rate dropped. In fact, in 2007 the rate for the lowest quintile was less than half of what it had been in 1983.
So, in the 25-year span from 1983 to 2007, the only quintile whose tax rates rose was the highest quintile, while the quintile whose tax rates dropped the most, by far, was the lowest quintile.
That’s looking at total federal taxes, which is Leonhardt’s preferred measure, not just income taxes. In terms of income taxes alone, the rates for the bottom two quintiles fell by more than 100 percent (to negative rates) across the most recent quarter-century that the CBO has measured, while the top quintile’s rates slightly increased. In all, the top two quintiles paid 99 percent of the nation’s income taxes in 2007, the top quintile paid 86 percent, and the top 1 percent paid 40 percent. The bottom three quintiles collectively made 25 percent of the nation’s total income but paid only 1 percent of the nation’s total income tax, according to the CBO.
The CBO’s figures stop before the recession, but the somewhat left-leaning Tax Policy Center has estimated numbers through 2009. Meanwhile, Leonhardt’s starting point of 30 years ago takes us back to before the Reagan tax cuts, which spawned about a decade’s worth of prosperity. But even if we use the Carter-era starting point of 1980, and the Tax Policy Center data for 2009, there is still no evidence to support Leonhardt’s claim.
For the 30 years from 1980 to 2009, according to these figures, total effective federal tax rates fell the most for the lowest quintile (more than 100 percent, into negative territory), then the second-lowest quintile (53 percent), then the middle quintile (28 percent), then the second-highest quintile (20 percent), and finally the highest quintile (16 percent). It’s true that rates for the top 1 percent fell by more (25 percent) than they did for either of the top two quintiles as a whole, falling to a rate of 26 percent (compared to -1, 7, 13, 17, and 23 percent for each of the five quintiles, in ascending order of income). But that paled in comparison to the declines among the bottom two quintiles.
What is more, the results for income taxes alone (as opposed to the results for all federal taxes paid) provide even less support for Leonhardt’s claims.
No matter how you slice it, the biggest beneficiaries of changes in tax rates over the past 25 or 30 years have been the two lower-earning quintiles—particularly the lowest-earning quintile—not the wealthiest quintile or the wealthiest 1 percent.
So on what is Leonhardt basing his claim that “Over the last 30 years, rates have fallen more for the wealthy, and especially the very wealthy, than for any other group”? It beats me.