Debunking Conservative Tax Sophistry


Posted by Bob Lord

I happened upon a post by Daniel Horowitz over at Red State that trotted out an old conservative trick, which is to refer to the ratio of the percentage of total income taxes paid by those at the top to their share of the total income. Because, as one would expect, that ratio, which I'll call the TTIR (tax to income ratio) is significantly greater than 1, the conservatives view it is proof positive that the wealthy already are paying their fair share. If you have two minutes, please read Horowitz's post here before continuing, but you should be able to follow me without doing so. 

Horowitz's logic is based entirely on totally unremarkable statistics, which he reports as real eye openers. After complimenting the "extraordinary work" of the Tax Foundation, Horowitz expounds:

So what about those rich one-percenters?  Well, in 2010, with the full force of the Bush tax cuts, and including all the so-called loopholes and deductions, they paid 37.4% of federal income taxes, even though they only earned 18.9% of the gross adjusted income in the country…What about the super-duper rich – the top 0.1%?  They paid 17.8% of the pie, while earning 9.24% of the AGI.

You've seen this argument and have already seen the standard counter-arguments. You know the argument conveniently fails to include all taxes in the analysis, because employment taxes and sales taxes both are regressive. But put that aside. You know that the adjusted gross income of the top 1% is a misleading number, because it doesn't include accretions to wealth quite common to those at the top that aren't included in AGI. But put that aside as well. 

Horowitz's argument — that this TTIR data demonstrates the rich are paying their fair share — is a complete crock. The TTIR of the top 1% bears no relationship to the fairness of the tax system. That TTIR almost is entirely a function of the distribution of income. Furthermore, the relationship between the TTIR and the distribution of income is parabolic, not linear. Okay, that was a bit geeky, but consider the following:

If the distribuiton of income were completely even, the TTIR would have to be 1.00 for any group, regardless of how fair or unfair the tax rates are. The top 1% would pay 1% of the total tax and would receive 1% of the total income. At the other extreme, if the distribution of income were completely uneven (that is, all the income went to the top 1%), the TTIR also would have to be 1.00. The top 1% would pay 100% of the total tax and would receive 100% of the total income. So, what drives the TTIR is where the distribution of income is along the continuum between completely even and completely uneven. Somewhere in between those two endpoints, and not necessarily at the mid-point, the TTIR maxes out. If you're at that point, the TTIR decreases if you change the distribution of income, regardless of whether you make it less even or more even, and when you get to either extreme the TTIR must be 1.00.

I tried this with a hypothetical population and it checks out. I took a population of 81 families and assumed the first 80 each had adjusted gross incomes of $50,000 and taxable incomes of $25,000, and the last family had an adjusted gross income of $1,000,000 and taxable income of $750,000. Under that scenario, the TTIR for the rich family was 2.538. I then made the income distribution less even by redsitributing $1,000 of income from each of the first 80 families to the rich family. That caused the rich family's TTIR to increase to 2.541. But when I made the distribution even more uneven by taking an additonal $2,000 from each of the 80 middle income families and giving it to the rich family, the rich family's TTIR decreased to 2.477. Thus, the TTIR peaked after the first redistribution, indicating I had found the point in the continuum between completely even income distribution and completely uneven income distribution where the TTIR maxed out.

So, what does this tired old conservative argument tell us?  Precious little, really. The statistics Horowitz cites in his post suggest that the distribution of income in America happens to be the distribution that causes the TTIR for the top 1% to be about 2.00. That's rather unremarkable. We have a tax system where the rate starts at 10% for all income above a statutory minimum, jumps to 15% very quickly, and maxes out at 35% for income above 388,000 for a married couple. Nor is it remarkable that the TTIR for the the top .1% also is just under 2.00. I have news for Horowitz and the other conservatives who push this argument. You could raise the top rate to 40 or lower it to 30, and the TTIR of the top 1% would barely budge. But if you redistributed the income of Americans to a 1970 income distribution, the TTIR for the top 1% would change significantly. And, by the way, if the TTIR for a subset of the population is a measure of the fairness of the tax system as applied to that subgroup, why is it larger for the top 1% than it is for the top .1%. That's a bit counter-intuitive, isn't it?

Do the statistics Horowitz and other conservatives rely on tell us anything about the fairness of the current tax system? Not at all. The fairness of the system can't be tied to the then current distribution of income, but the statistics upon which these conservatives rely are. Are they taking the position that the system they claim is fair would become less fair or more fair if economic conditions caused the distribution of income to change? I'd think not. 

The bottom line is that Horowitz's piece in Red State, which echoes a specious argument conservatives have been making for years, adds nothing to the tax debate. As countless other conservatives do, he uses statistics that are a function of the distribution of income in America to argue that our current tax rate structure is fair. The logic is absurd and the argument is baseless.