Bryan Caplan  

40 Years on the Status Treadmill

Krugman slides deeper into old... The European Union versus the ...
The General Social Survey has spent four decades asking Americans about their self-perceived status:
If you were asked to use one of four names for your social class, which would you say you belong in: the lower class (=1), the working class (=2), the middle class (=3), or the upper class (=4)?
During this period, Americans experienced substantial absolute gains in income and education.  Yet average status actually has a slight downward trend.  Here's what you get if you regress status on year.  (yeara=year/1000)


If you correct for rising income and education, though, status has noticeably fallen.  Translation: People today need higher levels of achievement to feel superior to other people.  See what happens as soon as you adjust for log family income, years of education, and attained degrees:


Magnitudes?  The last equation implies that from 1972-2012, achievement-corrected status fell by .14.  That's larger than the status gain people get when their income doubles.

Is this all obvious?  To me, yes.  But lately several economists I know have challenged me when I claimed that status is basically a zero-sum game.  For America during my lifetime, they seem to be wrong - and I'm here to collect a little of their status. :-)

COMMENTS (13 to date)
Steve Reilly writes:

If status has noticeably fallen, aren't you wrong too?

Mico writes:

One possibility is that the status attached to these class labels has also fallen. In the UK for instance 'middle class' historically meant something like the top 5-10% of the population. Now it is slowly coming to mean the middle 60-80%, like in the US.

BC writes:

@Mico, if status of these class labels has fallen, then wouldn't people self-rate themselves higher? If historically middle class meant top 5-10% and now means middle 60-80%, then presumably someone in the top 5-10% that used to self-rate as middle class would now self-rate as upper class.

Regarding Brian's point that "People today need higher levels of achievement to feel superior to other people," that would seem to imply, contra middle class stagnationists, that people perceive broad-based economic gains across the population as a whole. If people have "experienced substantial absolute gains in income and education" yet perceive their own status as falling, then they must perceive others' income and education as rising even faster than their own.

Chris B writes:

I wonder if the survey itself is flawed because it doesn't have a category "upper middle class". It was a decade after the survey started, in the eighties, that we started to see the rise of the mass affluent consumer, yuppies, etc. There is a vast gulf between middle class and rich. There is a much tighter spread between Lower Class, Working Class, and Middle Class. If you included an "Upper Middle Class" category, I would think you would have seen an overall status gain.

AS writes:

This reaffirms the need to be skeptical of statistics. It would be nice to see the raw data so we can see what's really going on. Four arbitrary buckets of "class" whose meaning is ambiguous and may not necessarily be collinear hardly proves a correlation. A more interesting question would be to ask people what percentile of class they believe they belong to.

Also it doesn't correct for immigration. If everyone who was in the US since 1972 feels better off, and it's only new immigrants who pull the stats down, then that can be a Pareto improvement even though the average has fallen.

Tom West writes:

It would be interesting to see how these changes occurred on a region-by-region basis.

My conjecture would be that better communication allows large pockets of somewhat poorer areas to compare themselves with the country at large, rather than their neighbors. What seemed like a middle-class neighborhood to its inhabitants is, nationally, now clearly a working class or even lower-class neighborhood. Their status drops, along with the accompanying happiness.

I see this with my sons. They have some strong talents, but instead of being the best in "X" of everyone they know, they can YouTube and instantly realize they're not a patch on the world champion. Status for their ability instantly drops from significant to only slightly above 0.

Of course, it also has farther reaching consequences. Resentment of both one's own circumstances and of the United States grows as, rather than the USA being some distant legend, you get life in the USA along with its incredible wealth beamed into your home every day, dropping your status (and your happiness) like a rock.

Hazel Meade writes:

I don't think that list of classes really encompasses the full spectrum of possibilities.

What do you do about people with a high level of education but low income or few assets?

I think there might actually be TWO middle classes:
1. Low income / high education
2. High income / low education

There's definitely a lot of people that fit in both categories. And both of them probably suffer a lot of status anxiety because of the mismatch between their education and income.

Eric writes:

The general point may well be correct, but the methodology is woefully inadequate. What makes you think that the status differences between each of these categories are equal and constant over time? That is required for OLS to be a valid.

Treating ordinal, cross sectional data cardinally is generally a bad idea, here and elsewhere (see: education research using achievement in "standard deviation" units).

dz writes:

Yes! The zero-sumness of status, combined with the very high diminishing returns to money, is *the* reason that wealth redistribution is such a net win for society. Because so much of the pie increase one can get via low taxes has the form of zero-sum status goods (iPads, fancy cars, etc.), it's worth upping tax rates and shrinking the pie *a lot* to make sure that everyone has the important non-status-based goods like food and shelter.

Bryan, I know you're not a utilitarian and don't necessarily care about things like "the total happiness of society", but I'm curious whether you agree that your views on status imply that high taxes on the rich are a good way to raise overall utility.

MingoV writes:

Years of propaganda have put a stigma on being upper class. I strongly suspect that upper class people are downgrading their status to middle class because they don't want to be among the top five percenters who are vilified by the media and politicians (except when they want contributions).

Matt writes:

NFL football games are also zero-sum competitions, but their existence generates many positive-sum transactions.

Jay writes:

I think Minko is right on. Admitting you are upper class is like admitting you are a racist. There is a very real social stigma to being upper class.

I remember years ago (~2000) in an intro econ class the professor asked people to self-identify their families as lower, middle or upper class. Nearly everybody identified as middle class. He then went on to ask for the family income.

It ranged from students like me who had family incomes of under $30,000 to students whose parents were physicians and business owners with incomes well over $500,000 a year. I don't remember the point of the exercise. But in hindsight, the point of the exercise could easily have been that politically hijacked language is ineffective when when trying to discuss economics.

Steve Sailer writes:

I agree that it's unfortunate that the GSS didn't start with a more expansive scale. The upper middle class is a very real phenomenon.

And there's no end to class divisions. For example, it took me years to understand the President's class background: he comes from the kind of Yankee-internationalist family that provides a lot of Foreign Service officers and Ford Foundation staffers, the slightly down-at-the-heels educated families that provided the State Department with its Arabists and Old China Hands.

He actually fit in very well at his Honolulu prep school founded by New England missionary Hiram Bingham I.

His best friend at Occidental and Columbia (they came up together with the plan to transfer) was the son of American foreign service officers. Most of his other friends at both colleges were Marxist scions of wealthy Pakistani families. His serious girlfriend Genevieve was the daughter of the #2 man in the Australian version of the CIA who later became Australia's ambassador to the United States.

His African father had a Harvard M.A. His stepfather came from the family of the top petroleum geologist in Indonesia. Granted, his American grandparents didn't have college degrees, but they were the black sheep of cultured Jayhawker families. His grandparents passed up college for war work and early marriage, but both his grandfather's brother and his grandmother's sister earned doctorates, as did his mother.

That's an unusual but not unknown background (especially among diplomats and international NGO officials). If you had to put one label on Obama's class while growing up, it would have to be "upper middle class." He never had the money to be "upper class" and his life up through his mid-20s was remarkably divorced from the "middle class."

Comments for this entry have been closed
Return to top