I wrote a short piece on the political untouchability of the upper-middle-class and why it’s a problem. My basic argument is that
(1) raising marginal tax rates is a terrible way to raise revenue, because it creates work disincentives;
(2) curbing tax expenditures are a much better way to raise revenue;
(3) but (2) threatens the interests of upper-middle-income households, particularly those residing in high-tax jurisdictions in high-cost metropolitan areas, a constituency that tilts to the left but that is important to both of the major political coalitions;
(4) and I argue that the collective political influence of the upper-middle-class is greater than that of the ultra-rich.
This last point should be obvious. The United States may indeed be a Plutonomy, in which the influence of the top 0.01 percent is enormous. But the influence of the slice of the population from, say, the 90th percentile to the 98th percentile is expressed in many different ways: small dollar donations, direct involvement in campaigns, and, most importantly but far more difficult to measure, through disproportionate cultural influence, i.e., heavy concentration in the upper echelons of the “culture industries,” including the news media.
An obvious check on the influence of upper-middle-income households is the ideological heterogeneity within this group: the upper-middle-class is not a monolith. Yet on an issue like tax expenditures, it may as well be a monolith, united across party lines by a shared interests in defending the mortgage interest deduction and the state and local tax deduction, among others.
To illustrate why the upper-middle-class matters in the tax debate, I recommend taking a look at this chart, which I found via the Hoosier Pundit. It measures the amount of total taxable income for all filers by adjusted gross income level for 2008, and it shows that the total taxable income for households earning $100,000 to $250,000 — comfortably beneath the upper bound of the “middle-class” — approached $1.4 trillion.
"raising marginal tax rates is a terrible way to raise revenue, because it creates work disincentives"
This is false.
Recall the experience of the 1990s, when we bumped rates up on income earned over $250,000. The economy boomed, the government had surpluses, happiness abounded as the country cheered. You're right that putting the top rate at 80 percent would probably have bad effects (though, of course, Eisenhower had much higher rates and a generally well-performing economy), but no one's talking about rates at anything like those levels.
"(4) and I argue that the collective political influence of the upper-middle-class is greater than that of the ultra-rich."
This is debatable; it is perhaps entirely false, but I don't think we have enough data to be sure. See this study: External Link
“Using an original data set of almost two thousand survey questions on proposed policy changes between 1981 and 2002, I find a moderately strong relationship between what the public wants and what the government does, albeit with a strong bias toward the status quo. But I also find that when Americans with different income levels differ in their policy preferences, actual policy outcomes strongly reflect the preferences of the most affluent but bear virtually no relationship to the preferences of poor or middle-income Americans.”
Now, that's one study, so it doesn't end the discussion forever. In all likelihood, though, it identifies a tendency in our politics.
Whether that makes us happy or makes us sad probably depends, for most of us, on our political preferences. If we want more protectionism, government programs advancing Christian and English-only policies, higher taxes on the wealthiest, and less recognition of gay relationships (or a return to imprisoning gays), then that lack of responsiveness to the views of the bottom 90% probably makes us sad.
Given that few people have all those views, it's probably a mixed bag. So then we get down to the meta question of whether it's a bad thing that policymakers are more responsive to the wealthiest.
Anyway, I don't think that your argument in this column holds up, because those premises are false or questionable.
Reply to this commentLinkReport AbusePlease scroll back through the archives of this blog re: MTRs.
Reply to this commentLinkReport AbuseThanks for reading and replying! Will go through the blog for your thoughts on marginal tax rates later on. Just because I just came across it, here's a few charts on MTRs & growth. They don't prove anything, in my view, but it's something to think about: External Link
Reply to this commentLinkReport AbuseUm, you are mistaken. The upper bounds of "middle class" are exceeded long before you get to $250k. At that household income, the median household income is exceeded by 4 or 5 times. The real world is a lot closer to $5k per month, not $500 per day. File this under the "everyone thinks they are middle class" error in logic box.
Reply to this commentLinkReport AbuseThe AGI measure doesn't capture capital gains and dividends, which is the main source of income for the "upper class". The share of the AGI contolled by the "upper middle class" is therfore a misleading measure of their political clout. That being said, I think Reihan is right that curtailng "upper middle class" tax expenditures will be very difficult--especially because the "upper class" will be perfectly happy to unite with the "upper middle class" to oppose such policies.
Reply to this commentLinkReport AbuseThat chart's a horrible way to present that data. What does the x axis mean? What do the bar widths mean (it's not equal slices of the population)? All the chart's really saying is that a lot of people make 100-200k. Going from a $25k/year bin (75-100k) to a 100k bin (100-200k) makes it look like there's a huge spike in the amount of money available at that range. There are more people who make 100-200k than 200-500k, so you see a sharp drop. A much better method would be to estimate the distribution of incomes from available average data, scale it by income, and plot it on a linear scale. It's not rocket science, but I guess it's too much to ask from someone who thinks Obama had much to do with spending in FY2010 and that the poorest 50% of Americans pay no income tax (payroll taxes are income taxes).
I think the point stands that you're not going to raise a ton of revenue by exclusively taxing our new, strange definition of the upper class; there's less than $2 trillion there and a lot of it's capital gains and not regular income.
Reply to this commentLinkReport AbuseYour claim 1 is an elementary category error "raising marginal tax rates is a terrible way to raise revenue, because it creates work disincentives;" You draw a conclusion about severity "terrible" from a claim about existence. What if the elasticity of labor supply were
0.000000001 ? Would the consequences of work disincentives still be terrible ?
If you wish to evaluate the relative costs of raising marginal rates and of broadening the tax base, you must be willing to deal with numbers, quantitative estimates. If you refuse to deal with quantities and try to reach conclusions using only English, you might as well just flip a coin.
I think I can assure you that your claim is nonsense. The reason is that estimated labor supply elasticities are quite low. I would refer to Martin Feldstein (totally against higher marginal tax rates, at the right limit of reality based public finance and very energetic). Absolutely the more nearly convincing arguments for low marginal tax rates concern tax avoidance -- tax shelters, contracts written to disguise salaries as capital gains or corporate profits etc etc.
The people most eager to make your 1) who actually deal with facts manage to reach the same conclusion but not via that argument.
What is your estimate of an elasticity of labor supply ? On what data is it based ?
Reply to this commentLinkReport AbuseI'm with Bob Simons. It makes me vomit in my mouth a little (even though I knew a lot of people thought this way) to read that the group between $100-250k is "comfortably beneath the upper bound of the 'middle-class'". I mean, come on: 80% of households (not individuals, but households) in the U.S. have incomes below $100k; only 5% have incomes higher than $180k!
(Source: Census Bureau, 2009)
Yet $250k (and on up, apparently, to $300-400k) is "middle class"?!? Even leaving aside the moral odiousness of this outlook, what about some basic mathematics, or semantics even? How can something up in the 97th or 98th percentile be in the "middle"? Using the same logic, then, a household with one part time minimum wage worker heading it would be "comfortably above the lower bound of 'middle-class'". Pffft.
I also agree with Zach about the graph. It might be more useful if all the bars were the same height, meaning the span of income each bar represented would be adjusted to make that be so.
Reply to this commentLinkReport Abusereflectionephemeral: You admit that large increases in marginal rates would decrease incentive, yet you deny that small ones would. Interesting.
Regarding the 90's, the proper way to look at it would be that the small increases in marginal rates were not sufficient to over ride the boom that was otherwise occuring in the economy.
You have not proven that small increases in marginal rates have no impact on incentives, just that small increases have proportionally smaller impacts.
Reply to this commentLinkReport Abuse