Update: More on Taxes and Class Warfare

Earlier this week I posted my thoughts on taxation, which included thoughts on taxation and class warfare and linked this recent WSJ editorial on tax shares paid by the rich.

Today, Kevin Drum rebuts the WSJ editorial with a post of his own.  Though I find Mr. Drum's consistent socialism and the-rich-will-be-first-against-the-wall rhetoric tedious, he is a smart guy and does have a point.  There is, as usual, a mixed message in the data.  However, this also means, as I will point out in a second, that Drum is guilty of picking and choosing his data points just as much as does the WSJ.

Drum points out, rightly, that while the share of taxes paid by the "super rich" (his term for the top .5% of income earners) has increased, their share of income has increased faster, such that their rates have gone down (god forbid that anyone violate the left's rule of the tax ratchet that says that tax rates may always go up but can never ever come down).  Using the same study as the WSJ, he rebuts this table of share of tax burden...

Share of Taxes (Income & Social Security) Paid By Income Classes

Category of Earners



1999 (at 2003 rates)

Top .1%




Top 5%




Top 20%




Bottom 20%




...with this chart , including Drum's subtle annotations in red:

His point is that the Super Rich actually pay lower rates now and the middle class pays higher rates, or as he puts it:

So shed no tears for the super rich in America. Their incomes have tripled in
the past couple of decades and at the same time their tax rates have decreased
by 9 percentage points. That's a pretty sweet deal in anybody's book.

Here are some thoughts on Drum's rebuttal:

Drum cherry-picked data too:  I will get back to the folks in the top 1 percentile in a minute.  Leaving them aside for a minute, note that Drum's storyline breaks down for everyone else.  If you compare the merely rich in the 1-20th percentiles, they got a smaller reduction in the Bush tax cuts than anyone in the middle and lower quintiles.  For example (comparing the 1999 before tax cut and 1999 after tax cut lines) the 1-5% richest got a rate reduction of 0.21%, while Kevin's favored group at 40-60% got a 1.45% rate reduction.

Don't blame this administration for previous tax increases:
  Drum is correct in saying that the tax rate has risen for the middle class over 20 years, but incredibly disingenuous not to explain why.   Note that the 1 point rise (which presumably Drum wants to hang on the current administration) actually consists of a 2.5 point rise from past tax increases, AMT creep, and payroll tax changes (passed by Democratic Congresses and generally supported by Drum) offset by a 1.5 point cut courtesy of the current administration.  Drum is in fact using data that clearly disproves his ongoing "tax cuts for the rich" mantra.  By the way, it is also interesting to see a good "progressive" ignoring progress on the lower two quintiles to decry higher taxes on the upper middle class -- seems like an interesting shift in focus.

Payroll taxes skew the picture:
  Including payroll taxes (social security and Medicare) in these numbers causes funny things to happen.  Why?  Because social security tax is straight-out regressive since it is flat up to about $90,000 in income and then zero after that.  This means that the total tax rate shown for the lower quintiles will include nearly 8% for payroll taxes (if this looks funny to you because it seems to imply that the lowest quintile must be paying negative income taxes, you are right, they are paying negative income taxes via the EITC).  However, as incomes rise above $90,000, taxpayers get an effective total rate reduction.  For an income of $180,000, a taxpayer only is effectively paying 3.1% to Social Security.  At $1 million, they are only paying 0.56%.  So, even if income tax rates were perfectly flat with no deductions for anything, those in the 1% category of richest people would have a total rate including payroll taxes over 5.5% points lower than the middle class.  If you recast the numbers above leaving out payroll taxes, you would not see the decrease in rates into the 1% group, the numbers would continue to increase, as can be seen here (from government data):

Effective Income Tax Rate (excludes payroll taxes) by income class

Category of Earners

2005 Fed Income Tax rate (effective)

Top 1%


Top 5%


Top 20%


2nd quintile


3rd quintile 4.1%
4th quintile 0.6%
Bottom 20% -5.6%

So, for income tax rates, there is still progressivity all the way to the top.  If you want to argue Social Security taxes, fine, but don't use Social Security tax effects to make a point about income taxes

By the way, in terms of the regresivity of Social Security, the defenders of that program need to stick with a story - is it a transfer payment or is it a government run insurance program?  If it is a government run insurance program (as defenders want to argue, since that seem more palatable to the public) then the $90,000 income cutoff makes sense:  Since the program does not pay benefits based on any incomes higher than this, "premiums" shouldn't be based on higher incomes.  Update: Kevin Drum says in this post that Social Security is

a modestly progressive social insurance program that's paid for by everyone and
that benefits everyone. If it ever stops being that, if it ever stops being
universal, it will eventually cease to exist.

OK, but stop lumping the "premiums" of this program in with income taxes to try to prove a point about the income tax system.

All that being said, there may be something funny going on in the top 1%:  As pointed out above, a portion of the apparent rate reduction for the top taxpayers is in fact due to the odd math surrounding Social Security taxes.  Any income tax cut, even if it is progressive, can make total taxes more regressive by shifting the mix to the very regressive social security tax. All that being said, the taxes of the very very rich are odd, because their income streams are so very different than those of you and I.   In particular, that weird mess of targeted tax reductions that I have decried on any number of occasions come much more into play in the very rich's tax returns, with results that are almost impossible to understand or forecast.  If Drum wants to use this data to argue for flat taxes and an elimination of deductions, I am all ears.

  • Thursday's Daily News

    It’s More than Judges - Larry Kudlow, National Review A Taxing Trend - David Keating, National Taxpayers Union The Senate’s Dr. Pain - Robert Novak, Townhall The Gas Tax Is Still a Terrible Idea - Chris Pope, AEI Diagnosing The...

  • Thanks for the good discussion with more light than heat (unusual in discussions of this subject). Based on the description at the link you cite at the CBO, the numbers in your table include the payroll tax and assign the imputed income (from both the individual and employer contribution) to income in the appropriate quintile and both the individual and employer contribution of the payroll tax to taxes paid cf.:

    The income measure used includes pretax cash income plus income from other sources. Pretax cash income includes wages, salaries, self-employment income, rents, taxable and nontaxable interest, dividends, realized capital gains, cash transfer payments, and retirement benefits; taxes paid by businesses (corporate income taxes and employers' shares of Social Security, Medicare, and federal unemployment insurance payroll taxes); and employees' contributions to 401(k) retirement plans. The comprehensive income measure also includes in-kind benefits (Medicare, Medicaid, employer-paid health insurance premiums, food stamps, school lunches and breakfasts, housing assistance, and energy assistance).

    Emphasis mine. Note that Medicare and Medicaid benefits are counted to the income side. So when this is taken into account the entire system is very mildly progressive (nearly flat).

  • Catching my eye: morning A through

    Here's what's caught my eye this morning: A Guy in Pajamas relates a conversation from a Japanese bar that I found rather interesting. It's hot girl-on-girl action as Ann Althouse comments on a Camille Paglia reading she attended last night....

  • The fact that the fourth quintile pay almost no tax and the lowest 20% pay a negative tax, raises the point of represntation without taxation. The lowest two quintiles have a strong incentive to vote for people who will raise taxes because it won't cost them anything (so they think) and it might benefit them. Higher taxes suppress prosperity and productivity so they hurt everyone. The Fair Tax (national retail sales tax) http://www.fairtax.org would be fair to all income levels and would give all citizens an incentive to keep taxes low. Because of the built-in "prebates" all taxpayers would be protected from taxation on all spending up to the poverty level. This is too small a space to explain all the many benefits of the FairTax but the web site answers all questions thoroughly.

  • The charge of "Taxation without Represenation" is unfounded in this case. The higher income distribution folks are very well represented in federal government. Even more so if you weight the percentage of goverment people by power (e.g. president and his cabinet are top 1% income earners, etc.)