Tax Fairness and Fairness in Tax Data Reporting

Did you know that the nonpartisan CBO recently demonstrated that the top 1 percent of American taxpayers "pay too much" in federal taxes? It was news to me, and no doubt to the CBO itself. We have Ari Fleischer to thank for this startling piece of disinformation.
This post was published on the now-closed HuffPost Contributor platform. Contributors control their own work and posted freely to our site. If you need to flag this entry as abusive, send us an email.

Did you know that the nonpartisan Congressional Budget Office (CBO) recently demonstrated that the top 1 percent of American taxpayers "pay too much" in federal taxes? It was news to me, and no doubt to the CBO itself. We have Ari Fleischer, President George W. Bush's former press officer, to thank for this startling piece of disinformation, via a late July op-ed in the Wall Street Journal.

Why pick on Ari Fleischer? After all, he is just a political operative, not a scholar or journalist. There are two reasons to rise in response to his op-ed. First, I want to offer the actual facts to those readers who have open minds. Second, I want to defend the honor of the CBO itself, since by virtue of its mission it is not able to respond directly to gross distortions of its work product.

The document in question is the CBO's July 2012 report, The Distribution of Household Income and Federal Taxes, 2008 and 2009. (It's online at This catchy title is actually a follow-up to an earlier 2012 publication with a similar title, but which covered the years 1997-2007.

CBO reports are written in a measured and slightly academic tone for careful and thoughtful readers, which may not serve the institution well when their documents fall into the hands of people like Ari Fleischer. To see my point, let's compare and contrast how Fleischer reads this CBO report with what it actually says.

Fleischer begins by asserting that the tax system is "incredibly progressive." Well of course it's progressive -- that's been the point of the federal income tax since its adoption in 1913. Fleischer reports, as if it is news, that those with higher incomes pay a larger share of the nation's total tax bill than do the poor (does he recommend the contrary?), and that those with higher incomes pay a larger share of total federal taxes than their share of total national income. Well again, that's just a definition of any progressive tax system.

But drilling down, what is the basis for the conclusion that the system is "incredibly" progressive?

To begin, let's observe that Mr. Fleischer consistently presents selective data relating to the tax burdens on those Americans in the top 20 percent of the income distribution. This group in 2009 had an average "market income" (of which, more later) of $218,000, which sounds pretty flush. But as the CBO Report carefully notes, "that average masks wide differences between subgroups [within the top 20%]; households in the 81st to 90th percentiles received $125,000, on average, compared with $1.2 million for households in the top percentile [i.e., the top 1%]." [CBO report p. 2.]

This point is critically important, because the current tax debate in Washington is not about how the top 20 percent of Americans by income should be taxed, but rather the top two or three percent. What Fleischer is doing is weaving an argument on behalf of the very affluent while hiding their actual tax and income data behind the skirts of a much larger class of reasonably but not extraordinarily successful taxpayers, whose tax rates are not in fact being threatened by the President or anyone else in Washington.

Fleischer elided over another critical point, which is that CBO's base measure of income is something called "before-tax income." (Here is perhaps the only place where I might gently suggest that CBO's standard presentations could use some tweaking to make it more difficult for other readers to make this mistake.) "Before-tax income" is not at all the same as "taxable income," which is what you pay tax on. "Taxable income" corresponds very roughly to what CBO calls "market income" -- your wages, net income from your own business, your retirement income and your investment income. CBO's base measure of "before-tax income" is the sum of your market income and the value of any transfer payments you receive -- not just "food stamps" (technically, the SNAP program), but also broad-based entitlements like Social Security payments, Medicare and Medicaid benefits, unemployment insurance, and much more. Most of these transfer payments are not includible in the taxable income of their recipients.

Needless to say, transfer payments are bottom-weighted, just as one would expect. Households in the bottom 20 percent, ranked by market incomes, received 40 percent of transfer payments (an average of $23,000 per household). [CBO report p. 2.] Almost two-thirds of that came from Social Security and Medicare, not some imagined "welfare" programs. [CBO report p. 2] The inclusion of transfer payments radically changes the percentage of national income claimed by different income groups. In particular, the top 20 percent of households in the income distribution in 2009 earned 50.8 percent of aggregate before-tax incomes, but 56.8 percent of market incomes.

Now, see what Fleischer is doing. He argues that the U.S. tax system is "incredibly progressive," by presenting data on how much tax the top 20 percent actually pay, and comparing that, not to their share of aggregate taxable incomes, or its rough surrogate, CBO's "market incomes," but rather to their share of national income after taking into account the very transfer payments that are funded by the tax system. By doing so, he makes the average tax rate shouldered by the top 20 percent seem higher.

If what you want to do is ventilate about the progressivity of the tax system, you want to compare taxes paid to market income, not before-tax income. Otherwise you are chasing your own tail, since a principal purpose of a progressive tax system is to fund the very tax-funded transfers that CBO a bit confusingly labels as components of "before-tax" income.

CBO in this report further inadvertently helps Fleischer overstate reality by taking the refundable portion of the earned income tax credit (and other refundable tax credits) and presenting those refunds as negative taxes, rather than as income. [CBO report p. 3.] (This presentation is odd, because it conflicts with how refundable tax credits are presented in the federal budget.) This further drives down the aggregate tax burden (as presented in this report) on lower income Americans. Again, if you want to talk about tax burdens and progressivity, you should do so before taking into account tax refunds that exceed taxes paid, because those are transfer payments by another name.

Fleischer probably also did not notice that CBO presents its reports by reference to "households," which are not the same as "taxpayers." A husband and wife, along with another adult relative (like a mother-in-law) typically will constitute two taxpayer units, but only one household. (CBO also restates the percentage ranking of households through a complicated approach that attempts to neutralize differences in the cost of living for households of different sizes. CBO Report pp. 24-25.) So the income levels of CBO's "households" at a particular income distribution point are not the same as that of "taxpayers" at the same point in the income hierarchy.

Progressivity is actually a tricky thing to measure, because it is a comparative thought -- one system is "very" progressive only in comparison to another system. But in light of Mr. Fleischer's assertion that the CBO report demonstrates that the top 20 percent of Americans pay "too much," one might fairly begin with the obvious question, does the tax system tax the top 20% of taxpayers more harshly than it did in in 1979 (the beginning of CBO's data)? There the data are clear: the average tax rate for the top 20 percent of taxpayers (measured by before-tax incomes) "reached a new [modern era] low in 2009, about 4 percentage points below its 1979 level." [CBO report p. 15.] And as it happens President Obama presided over a 1.7 percentage point drop in the average tax rate of this group. [CBO report p. 15.] A roughly similar story applies to the top one percent crowd -- their effective tax rate is much lower than it was in 1979, and has declined steadily since 1994, subject to a very small uptick in 2009.

If you actually care about tax system progressivity, as opposed to tax propagandizing, then you might want to start with a standard measure of progressivity, like the "Gini index" (after a statistician named Mr. Gini, not a misspelled tribute to Barbara Eden). The higher the number, the more unequal a country's income distribution is. A similar concept is applied to tax systems by themselves; CBO calls that the tax progressivity index.

CBO crunched all the income and tax distribution inequality numbers, but unfortunately did not include them in this report. You can find them, though, in the supplemental tables to the CBO report on the CBO's website. (It's Table 9 of the Supplemental Data.)

What do these data actually say?

They tell the depressing story of a country riven by ever-increasing disparities in incomes. The year 2007 (right before the crash) marked the greatest inequality in the distribution of market incomes in the history of CBO's recordkeeping (dating back to 1979). Inequality marched steadily upwards year after year during that period, except for a brief interruption by virtue of the 2001 recession. And while transfer payments mitigate some of the inequality attendant on pure market outcomes, when one examines the Gini indices for both "before-tax incomes" (that is, market incomes plus transfer payments) and after-tax incomes, those figures also hit an all-time high in 2007. The current recession reduced the incomes of the very affluent disproportionately in 2008 and 2009, so that the Gini index measures of income inequality in those years were slightly off the 2007 high water mark, but the CBO report tells us that preliminary data for 2010 show that the affluent among us have already recovered much of their lost ground. (CBO report p. 16.)

Now let's consider, not the distribution of incomes, but the progressivity of the tax system in the abstract. The relevant tax progressivity index is the index that compares the progressivity of before-tax and after-tax incomes. (In my view, as stated, it should be a comparison of market incomes and after-tax incomes, but CBO did not calculate that.) CBO also offers an additional measure that is largely meaningless, which is the percentage of the total tax bill paid by different income levels. If the entire tax system collected $1000 in tax revenues, total, and all of that from one man, this alternative measure would suggest that we have a hugely progressive tax system, when in fact 330 million of us would pay nothing.

So here's the big punch line -- how progressive in fact is the federal tax system, according to the CBO data that accompanied the report that Fleischer asserts proved that the rich pay too much? Well, the 2009 tax progressivity index (which again compares before-tax and after-tax incomes to see the "leveling" effect of the tax system) was at exactly the same level as it was in 1979 -- and what's more, from 1994 to 2009 that measure of progressivity hardly fluctuated at all (except for 2007, when the tax system did somewhat less to reduce inequality than in the surrounding years). The tax system has not gotten more progressive in any meaningful sense, and at the same time the rich in fact are pulling away faster and faster from the rest of America, whether measured by market incomes, before-tax incomes, or after-tax incomes.

As the CBO patiently explained in its report, what actually went on in 2008 and 2009 has nothing to do with the story Fleischer wants to tell, and everything to do with the depths into which the economy fell. Essentially, the whole income distribution curve shifted dramatically to the left (towards zero, not towards Democrats): the very affluent made a lot less money, but still earned a lot, while millions of Americans lost their jobs and saw their market incomes plummet to near zero.

As one result, total federal tax collections as a percentage of total national income fell to a modern era low (CBO Report, p. 11), and every segment of society, including the top 20 percent, paid a lower average tax rate than they did in the boom year of 2007. (The exception was the top one percent's tax rate; these fortunate individuals actually saw their average tax rate go up a little bit, because a smaller percentage of their total income came from capital gains, which is taxed at a preferential 15 percent rate.) (CBO Report p. 15.) As another consequence, if millions are suddenly rendered jobless, of course their tax bills go down, because they have no incomes on which to pay tax. The result is that the much smaller total tax bill for the country looks like it is being paid disproportionately by the affluent, but again that is a meaningless statistic, because it is an artifact of total tax collections collapsing as a percentage of national income: as a percentage of their incomes, the affluent (at least below the top one percent) paid a lower tax rate, not a higher tax rate, in 2009 than in 2007.

But there's more. The years 2008 and 2009 were exceptional in many other respects as well. Most important, in early 2008 President Bush signed the Economic Stimulus Act of 2008, which gave taxpayers bonus tax refunds of up to $600. (In signing the law, President Bush commented, "Congress passed a really good piece of legislation.") CBO in this report treated these bonus refunds as negative taxes, not additional transfer payments. (CBO Report p. 3) The same was true for emergency "making work pay" credit in 2009 of up to 2009, along with other recession-fighting tax measures. These bonus refunds were a small matter to the affluent, but greatly reduced the percentage of taxes paid by lower income Americans in those years. (CBO Report p. 3.)

In addition, CBO decided in this report to make a fundamental change in how it calculates the value for transfer payment purposes of Medicaid benefits; this accounting change in the valuation of Medicaid benefits ("accounting," because it changed nothing about the government's actual costs or recipients' benefits) caused CBO to show lower-income Americans as having $200 billion more in "before-tax income" in 2009 than would have been true under the old methodology. (CBO report p. 20.)

So what happened in 2008 and 2009 is that incomes collapsed, and because the tax system is progressive, higher income Americans still had some positive tax liability, while millions of less fortunate Americans dropped temporarily off the income tax rolls. This apparent concentration of tax responsibilities was exacerbated by all the emergency measures adopted under both Presidents Bush and Obama in 2008-2009, because those benefits were either per capita or bottom-weighted in their distribution.

As a result, in 2009 the affluent were asked to swallow a larger share of the radically smaller tax pie than was the case in 2007, when the tax liability pie itself was larger. But either way, the real question is, how much tax pie did I have to swallow? And in this respect, the affluent (other than the top one percent), like other Americans, in fact paid less in tax as a percentage of their income than they had in the past.

Finally, Fleischer notes that the CBO report includes both payroll and income taxes, but overlooks (or fails to state) that the report carefully notes that its figures include not only taxes paid by or on behalf of individuals, but also corporate income taxes, which the report apportions to individuals predominantly on the basis of how much capital income they earn. This means that wealthy taxpayers are credited not only with the taxes they actually pay, but also with the lion's share of corporate income taxes. Of course individuals ultimately bear the burden of the corporate income tax, but precisely which individuals do is hotly contested. I find it ironic that those aligned with Mr. Fleischer's political views typically show great flexibility of mind on this conundrum, by arguing that corporate taxes actually fall on the wages of humble workers when those individuals are proselytizing for lower corporate tax rates, and then claiming that the taxes clearly are a burden to the wealthy when the question shifts to the tax indignities forced upon America's highest income families. Keep this in mind when in a future op-ed Mr. Fleischer argues that he is trying to do the working man a favor by reducing corporate income tax rates.

Popular in the Community