The top one-hundredth of one percent

Alan Reynolds
Posted: May 11, 2006 12:05 AM

The fifth in a series of Washington Post editorials lauding the political reshuffling of income begins on the wrong foot. "The quest for ways to reduce inequality," the editors wrote, "begins with taxation. Unlike spending programs, redistribution through taxation is administratively simple."

But collecting taxes is not as easy as it sounds. And taxes don't redistribute income -- they just reduce income. Means-tested federal transfer payments account for little more than 10 percent of federal spending, and more taxes won't change that because the poor don't lobby or contribute to campaigns and rarely vote.

The editorial opines that "it would almost certainly be safe to increase taxes on the top 1 percent by 5 percentage points, restoring the level of the mid-1990s -- hardly a period of lethargic chief executives. ... To remedy stagnant middle-class living standards, more radical tax hikes would be necessary."

More radical did not mean toying with unsafe tax rates on 1 percent -- it meant casting a wider net. As Washington Post columnist Stephen Pearlstein wrote in March, "The classic redistribution scheme would be to raise taxes on the top 10 percent of income earners, roughly those with household incomes above $125,000."

The top 10 percent already pays two-thirds of the income tax, so a smart salesman begins with a foot in the door. The editorial started with that opening bid for adding 5 percentage points to the 33 and 35 percent tax brackets, yet quickly changed the subject to just the top 0.01 percent. That is one one-hundredth of one percent -- 13,359 taxpayers in 2000 who earned more than $5.3 million. The editorial, by contrast, proposes a 15 percent higher marginal tax rate on couples earning just a bit more than $200,000.

"According to Thomas Piketty of the École Normale Superieure in Paris and Emmanuel Saez of the University of California at Berkeley," says The Washington Post, "the top 0.01 percent of households has seen its tax bite fall by 6 percentage points since 2000 and by an astonishing 25 percentage points since 1980."

Whatever that astonishing diversion was all about, it was not about the individual income tax. The Congressional Budget Office (CB0) estimates that the average individual income tax rate of the top 1 percent was 22.3 percent in 1980 and 24.2 percent in 2000. The tax could hardly have fallen by 25 percentage points since 1980, when it was about 22 percent, unless we could imagine the top 0.01 percent paid twice as high a rate as the top 1 percent as a whole.

The CBO's estimate of the top 1 percent's share of the corporate tax was 10.8 percent in 1980 and 8.2 percent in 2003, so adding the corporate tax can't solve this mystery.

What makes the "astonishing" estimates so different from the CBO's is that Piketty and Saez toss in the estate tax and gift tax. That might make sense if all taxpayers in the top 0.01 percent died each year and left their businesses and lifetime savings to heirs with incomes as high as their own. In any case, the estate tax is absolutely irrelevant to The Washington Post's effort to promote much higher income tax rates.

There was a graph purporting to show the drop -- "in percentage points" -- of average federal tax rates by income group under President Bush. Taxpayer incomes were grouped into five quintiles (fifths), with the first meaning the lowest income, and the top 10 percent was sliced into slivers such as the top 1 percent and smaller.

To show what is fundamentally fraudulent about the "percentage points" trick, the table uses estimates kindly provided by Emmanuel Saez of average individual income tax rates for the middle fifth (quintile), the fifth just above that, and that miniscule "top 0.01 percent." There is no point showing average tax rates for the bottom 40 percent because their tax rates were below zero in 2000 and 2005 because of the increasingly generous Earned Income Tax Credit and (in 2005) the child credit.

Average Individual Income Tax Rates (Federal)

----------------------- 1980 ---- 1990 ---- 2000 ---- 2005
Middle 20 percent ---- 8.9 ----- 7.2 ------- 6.6 ------ 3.3
Fourth 20 percent ---- 12.1 ---- 10.2 ----- 10.3 ------ 7.5
Top 0.01 percent ---- 40.2 ---- 25.8 ------ 32.2 ----- 28.1

Source: Thomas Piketty & Emmanuel Saez

The table shows that those in the middle quintile saw their average income tax rate drop by 3.3 percentage points from 2000 to 2005. It doesn't sound like much in percentage points, but if their income was the same, their tax bill was cut in half. The next-highest quintile saw their tax rate fall by 2.8 percentage points -- a big 27 percent cut for the "upper middle class." The super-rich top 0.01 percent saw their average tax rate drop by 4.1 percentage points since the stock option peak of 2000, but that was not even 13 percent.

The table shows that by 2005, the average tax rate of the fourth-highest quintile was much lower than what the third-highest quintile had to pay in 1980. The average tax rate of the third-highest quintile in 2005 was likewise lower than that of the second-highest in 1980. The bottom two-fifths were taken off the income tax rolls entirely and usually given a cash subsidy if they worked at all.

Inexplicably, The Washington Post describes all that as a "loss of progressivity" and claims it was "at the expense of" somebody or other. Everyone, particularly non-suicidal Democratic candidates for political office, should take a close look at those 1980 tax rates (and the 1980 economy) before buying what some are trying so hard to sell.