U. S. Income Distribution—a Chart to Contemplate

by Dave Schuler on March 4, 2013

The chart above, courtesy of Census Bureau data, illustrates household income distribution in the United States based on the most recent information available. You can click on the chart for a larger version. I find this chart interesting for any number of reasons. Note that it illustrates in one compact form the distribution of income, median income, the five income quintiles, and the incomes of the top 10%, 5%, and 2% of income earners.

There are any number of observations I could make about this chart. I’ll limit myself to just four.

  • Why aren’t there more households to the left of the highest bar in the chart?
  • Why is the highest bar in the chart, representing 6% of households, the $15,000-$20,000 bar?
  • What’s with the two bars at the far right of the chart, the highest income earners?
  • How has the chart changed over time?

Last question first. The first time I saw a chart like this was a little over 40 years ago. I found it a bit shocking back then. It was actually pretty similar to this chart other than that the income numbers were all smaller. The biggest change I can recall is that those two bars at the far right are both higher than they used to be and, if I remember correctly, the several bars just to the left of them are lower.

Now, why aren’t there more households to the left of the highest bar? I think that’s an artifact. It’s been created by policies like the minimum wage and various policies that produce non-wage and non-taxable compensation, e.g. Medicaid. Below a certain level earning taxable income doesn’t make any sense since it lowers non-wage, non-taxable income. It also might be an artifact of reporting.

I think it’s pretty clear why the highest bar in the chart is where it is: essentially, that represents household with one wage earner, working full-time, earning minimum wage. They don’t comprise a vast number of household but there are enough of them to make it the highest bar. I also think that the several bars just to the right of that one largely consist of households with more than one wage earner, both earning minimum wage or, possibly, slightly above minimum wage.

That brings us to the real question. Why are those two bars at the far right of the chart so high? Beats me. I can’t think of any likely natural explanation. Note that the second bar from the right should actually be thought of as ten bars. Each of those bars would be quite a bit lower than that single bar but as depicted it neatlly illustrates a point I’ve made around here fairly often around here: you’ve got to think beyond the top 1% of income earners.

As I say although I can’t think of any likely natural explanations I can think of explanations. It isn’t intelligence. Intelligence isn’t distributed in the population that way and after you get beyond two standard deviations the correlation between income and intelligence just isn’t that close. The explanations I can come up with include inherited wealth and various other forms of rent-seeking. There are about 115 million households in the United States. 1% of that is 1.15 million households. 2% of that is 2.3 million households.

Remember that the tail on the right is actually very, very long. To my eye that still doesn’t explain why the second bar (actually ten bars combined) is as high as it is. Even if you divide it by ten it’s still too high as is the bar on the right which is actually many, many more bars.

There are roughly a half million physicians in the country. Nearly all of them are in those last two bars in the chart. My inference is that although they comprise a big chunk of those bars they don’t comprise all of them. Lawyers who graduated from the top 15 law schools comprise another chunk of those two bars as do middle and upper level managers of medium-sized and large companies and some top managers at small companies.

How much of what’s being illustrated by those two rightmost bars is the distortionary effect of our outsized financial sector?

{ 17 comments… read them below or add one }

PD Shaw March 4, 2013 at 2:59 pm

It would be interesting to remake this chart with the reporting of healthcare compensation on this year’s W-2 for total compensation, if that were possible. Would that create a slight bimodal distribution?

It seems to me that the two bars on the right are just limitations in the graph; they simply didn’t want to take it out in 5k units as the bars became imperceptible. I really hope there isn’t a “hump” somewhere past $200k that would tell us something interesting.

Dave Schuler March 4, 2013 at 3:03 pm

Would that create a slight bimodal distribution?

I think it almost certainly would.

As I wrote in the body of the post, I think that there’s no natural explanation for the second to the last bar, even divided among ten bars.

Icepick March 4, 2013 at 3:15 pm

Why aren’t there more households to the left of the highest bar in the chart?

A combination of EITC, SS and SSDI, at a guess.

Steve Verdon March 4, 2013 at 4:43 pm

It would be interesting to remake this chart with the reporting of healthcare compensation on this year’s W-2 for total compensation, if that were possible. Would that create a slight bimodal distribution?

Heh, I was thinking the same thing as well.

Drew March 4, 2013 at 5:16 pm

This issue has become one of fascination for me, because I think it is going to define politics over the next 20 years. (And I’m not optimistic, by the way.) I have no demonstrably proveable views or answers, like I do on the whole “Incentives” thread. Just observations and views that I hope might spur reasonable debate.

It is undeniable that incomes for the very talented/sainted/rare/desirable blah, blah, blah have increased disproportionately. But could we slice and dice that somewhat?

We have entertainment (music, movies, sports etc.) Historically the actual entertainers made jack squat. (I know, movie stars have always been movie stars, but there has been a sea change in entertainment comp structure) “Free agency” in sports, and of all things, Led Zeppelin in music changed all that. The money goes to the performers, although the top performers. I’m willing to make the value judgment that I think that is appropriate. What I’m not willing to do is say to The Average Joe that if they want to spend their hard earned money on stars they should be expecting tax subsidy from the wealthy or pity for not having money for groceries. Michael Jordan, P Diddy and Patrick Kane thank you for the ticket prices you pay, but don’t expect a care basket from them when you need a new car. Seems like a bit of a self inflicted wound.

The Professions

Law, finance, medicine, hired management. You all know my views on price discovery in medicine. steve would know better than I, but clearly the surgeons are very well compensated and in those upper tiers. But are primary care physicians? And are they not the bulk? And is the disparity between surgeons and internests warranted. My gut says no. But I’m no authority. In any event, its a “subsidized” industry based upon third party payers……….a system the Average Joe votes for. Self inflicted would #2.

Law and finance? For the most part one and the same, for the top 15 law school grads paper the finance…………at $600-$1200 per hour. Is finance subsidized? You bet. To paraphrase “how do you like me now?!” Who’s subsidizing them now? Obama. Who voted for Obama? Law and finance………….and the Average Joe who rails against Big Law and Finance. Self inflicted wound #3.

I know hired management in my segment of the market like the back of my hand. I will tell you that finding the right people is worth the weight in gold. Period. That’s a value judgment, but I will defend it to the death. Large corporate? Don’t know. But suspicious.

Business Owners
Anyone who wants to cure income disparity by penalizing business owners is a moron. They are the best of the best. (Business, not necessarily character) The biggest balls/risktakers. The street smart. The people willing to put it all on red. Deserving of the spoils. The people who make jobs for the Average Joe. Don’t make tax policy self inflicted wound #4.

In summ, I think the returns to value creation capability have become primary in the last 30 years (pulling a lever every 15 seconds on an assembly line doesn’t cut it), even if some is subsidy/rent seeking facilitated. I can’t imagine that it will be any less in the future. Fixing it through tax based redistribution is foolish and just self defeating. Reducing it through less government subsidy and rent seeking seems to escape Democrats worldview.

I’m not optimistic.

Andy March 4, 2013 at 8:04 pm

The average recipient of Social Security gets about $15k a year in income from the program. Coincidence?

In addition to what PD said, I would like to further break down the data to include the various types of income (wages, non-wage compensation, investment income, government benefits, etc.).

Eric Rall March 4, 2013 at 8:43 pm

To my eye that still doesn’t explain why the second bar (actually ten bars combined) is as high as it is. Even if you divide it by ten it’s still too high

Is it? Eyeballing the graph, the second bar is about 1.75%, and the last regular-width bar is about 0.25%. Ten bars the height of the last regular-width bar would add up to 2.5%.

There’s probably also a blip up at $200k (making it higher than the 195 bar), same as for $100k and $150k, due to our taste for round numbers.

TangoMan March 4, 2013 at 8:45 pm

What’s with the two bars at the far right of the chart, the highest income earners?

They’re aggregated columns ($200-$250) and ($250+) instead of income groups broken into $5,000 increments.

Drew March 5, 2013 at 9:27 am

I think we have analyzed the technicalities of the chart to death. Anyone have an answer to greater income inequality, or what to do about it?

Is it just a natural and unaddressable artifact of the way our economy has evolved the last 50 years?

Dave Schuler March 5, 2013 at 9:33 am

I think I’ve made my view clear. I don’t think it’s a “natural and unaddressable” feature of our economy. I think it’s an artifact of the web of subsidies we’ve put in place. It’s long past time to bring government out of the 1950s and start winding down the subsidies even as government’s consumer services are improved. We can rebuild it. We have the technology. We can make it better…faster…

Recipients of subsidies will fight their loss to the death. Indeed, their lives do depend on it.

The alternative will be catastrophic system failure and the outcome of that will be worse than if we wind the system down in a controlled fashion.

PD Shaw March 5, 2013 at 10:09 am

I think most of the differential here outside the top 1% is related to marriage. Look at this chart, showing how much the household incomes have diverged since 1950 on the basis of the marital situation. Its not simply that a two income household makes twice as much as a one income household — a married couple with one income will usually make more than a single person. Plus, the percentage of people married has declined over time, so the divergence has to be considered in light of the increasing percentage of those single.

BTW/ a family physician makes btw/ $130k and $200k, so he or she usually wouldn’t be in the top 1% unless there was additional marital income. My impression is that with more female doctors, its much more likely that doctors marry doctors.

PD Shaw March 5, 2013 at 10:14 am

Of course, I don’t have an answer for it. It seems like this is the natural result of women’s lib, not just in terms of greater economic/employment opportunities, but also because women are far more likely to consider the economics of a partnership than men and they are influencing the outcomes. Its just that a certain percentage of women are benefiting from these changes.

PD Shaw March 5, 2013 at 10:15 am

That final sentence would better read: Not all women have benefited from these changes.

Drew March 5, 2013 at 11:18 am

You know I couldn’t agree more with what you wrote at 9:33, Dave.

But I’m at a loss as to how to make it happen. I don’t know if you – or anyone here – shares my utter disgust with the media, but in my view it is 80% -90% in the tank for the left and an excuse making machine for them. Its a hard headwind to buck. And it will go to the wall to carry the water against an alternative, small government voice, choosing to go for the sob stories (you ain’t seen nuthin’ yet on the sequester) vs a sober view that yes, we can still provide for the poor and disabled, and deal with various necessary communal social functions through government and still bring government into the current century. Oh, hell, let’s at least get them to 1980. Its a start.

I think people underestimate the propaganda power of the cacophony of media. Short attention spans; having to deal with the kids; watching football… Just listen to man in the street interviews. But somehow, some way, we need to get rid of the patrician/rent seeking ways of government. Finances are at or near a breaking point.

Icepick March 5, 2013 at 2:33 pm

The alternative will be catastrophic system failure and the outcome of that will be worse than if we wind the system down in a controlled fashion.

Fuck the revolution, bring on the Apocalypse.

MJA April 4, 2013 at 3:40 pm

There’s a perfectly natural explanation for the 2nd bar from the right. And probably both. The 2nd bar from the right represents 10 income blocks. The total looks to be abt 1.75% of households, which is an avge of 0.175% households per block, less than the 3rd bar from the right, which looks to be at about 0.25%. I’m sure the same logic would apply to the last bar from the right as well. This chart is, after all, essentially a Bell curve.

participant-observer-observed January 26, 2014 at 1:09 am

The bump at the far right is probably from the dual citizen/green carder 1%s from other countries and their kids in US universities.

Leave a Comment

Previous post:

Next post: