Politically Incorrect Paper, a continuing series

Several years ago Bill Cosby chided poor blacks for spending their limited incomes on high-priced shoes and other items of conspicuous consumption instead of investing in education.  Cosby was widely criticized but I went to the numbers, specifically Table 2100 of the Consumer Expenditure Survey and found the following for 2003:

Average income of whites and other races: $53,292.
Average income of blacks: $34,485.

Expenditures on footwear by whites and other races: $274
Expenditures on footwear by blacks: $440.

As I noted then "to do a proper comparison we
would have to correct for income and other demographic variables."  The correction has now been done by three researchers in an NBER working paper (non-gated version).  The results didn’t surprise me.  How about you?

Using nationally representative data on consumption, we show that
Blacks and Hispanics devote larger shares of their expenditure bundles
to visible goods (clothing, jewelry, and cars) than do comparable
Whites.  We demonstrate that these differences exist among virtually all
sub-populations, that they are relatively constant over time, and that
they are economically large.

To give the authors credit where credit is due they also show that the differences in conspicuous consumption are large and important.  The differences in spending on clothing, jewelry, and cars, for example, can explain half of the differences in wealth between the races (conditional on permanent income) and a significant share of the differences in education and health spending.

Why do these differences exist?  Aside from simple differences in preferences, signaling is one possible explanation.   Suppose that high income confers status.  Other people judge your income based on your conspicuous consumption and your group’s income.  Under plausible conditions, the authors show that if your group’s income is already high conspicuous consumption has a low marginal product.  Put differently a black man who wears a very expensive suit gets a bigger increase in status than a white man who wears the same expensive suit because the baseline income prediction is lower for the former. 

The theory is plausible but I wonder if other groups haven’t converged on more efficient methods of signaling.  Some groups, for example, use education as a signal.  Other groups like to show how clever they are by writing pithy summaries of new economics research.


I'm a high school teacher in a school where 44% of our population is low income.
We study pooverty at every inservice. We rely heavily on the research of Ruby Payne.
Her conclusions match the findings in this post. Now, let me ask, is it more
important to look relatively good by buying expensive suits and toys because that
differentiates one or is it better to fit the stereotype that white middle class
think a low income earner should look like while saving for the future? Ms. Payne
also found that low income earners also value entertainment more than their upper
class peers.

When is white America going to catch up? How are we going to close the sneaker gap? As white Americans, are we going to just settle for another pair of drab NB? Or are we going to start to pick up the slack and do our share?

NB: Ruby Payne is wildly controversial in the education world (I take no position on her myself).

This post circles around, but never makes explicit, the point that whiteness itself confers status. For free.

Floccina: people wearing better clothes is a positive externality. People spending huge amounts of money on tasteless bling (14k gold spinners on your lifted, diamond-studded escalade) is a negative externality. Good clothes are a combination of taste and money. Just money doesn't result in good clothing. I would be interested to see a study that could determine if the additional conspicuous consumption engaged in by blacks was along the tasteful (must impress the white man so I can get a job) or the distasteful (must impress my peers so I can get laid) lines.

Sameer Parekh hit the nail on the head. You can look professional without spending a lot of money. A few nice looking suits can be worn for years, can be purchased for a few hundred dollars and can help you overcome the implicit bias towards being white (as one commenter suggested was an issue). Buying "rims" for your car or gold jewelry (which cost more than a few suits will, BTW) does not help you get a job in "White America" and is probably more about signaling to women that they should do dirty things to you.

This form of signaling is probably pretty universal. As JSK pointed out, it is just as important among the white population in the US (or in Germany, for that matter - some people spend outrageous sums on their cars). I'm new to this blog, and you might have already discussed Banerjee-Duflo's "Economic lives of the poor" (see http://econ-www.mit.edu/faculty/eduflo/papers ).
They find that the (extremely) poor -- less than $2 ($1) a day -- spend a good deal of their money on weddings, funerals, religious festivals, and literally starve to do so. That's an even more extreme form of signalling.

Finally, just a thought - the "causality" between spending relatively more on conspicuous goods and less on education may well go the other way round.

And some only have bandwith to write comments beneath those pithy summaries.

Many of the comment made here seem to indicate that many posters think that the majority of black income is controlled by teenagers and pimps, because this is the image that fit their stereotypes of Blacks. There is far more income controlled by Black office workers, salesmen, teachers, repairmen, bank tellers etc., who dress better that whites in corresponding jobs.

My wife came from a very poor upbringing says people who've never been truly poor don't understand how good it made her feel to have something new like everyone else, or even just a clean white t-shirt or non-ratty pair of jeans.

Regarding clothes: In my observations African Americans are generally more stylish than whites.

Signaling and conspicuous consumption are all around us regardless of ethnicity. I take issue with the apparent sloppy use of statistics.

The reported 'average' data are meaningless unless tested as statistically significant to some level of precision. Where are the "median" numbers? Perhaps I missed the statistical tests buried somewhere in the paper. I saw nothing on table 2100.

Merely being white is a way of signaling wealth. If you don't believe it, visit Peru.

I find this yet another example of Puritanism and racism masquerading as research. The unstated bias is that "conspicuous consumption" is morally reprehensible and thus by tying it to race we get the implied result that blacks are just irresponsible.

First, the definition of conspicuous consumption assumes the answer. If one restricts this to things like clothing and cars then you have biased the results. I don't see McMansions or swimming pools as being used as a factor. "Housing" is not the same thing.

Second, the data compares a group with a significantly lower income to one with a higher one. But the white population is not uniform, a more meaningful comparison would to have been to compare only those whites who matched the socio-economic status of the blacks. Then it would have been possible to find if spending on clothing was a racial characteristic or an economic one.

Third, as usual, the mathematics is worthless. Proposing "equations" where the symbols represent an unspecified function is completely meaningless. All that is needed to analyze the data are the standard statistical tests. I find none, except a best-fit straight line drawn through some of the data. There are no error bars, there is no measure of standard deviation and there is no measure of the correlation coefficient.

Eyeballing some of those graphs, one would be hard pressed to see any meaningful statistical relationship.

I don't know who NBER represents, but judging from a number of recent papers the quality of what gets posted under their aegis makes the whole enterprise suspect.

RobertDFeinman and a bunch of others get the story wrong. The authors introduce the signaling model to test whether the conspicuous consumption effect is due to preferences that just happen to differ by race or due to income. What they find is that relative income is doing most of the work. That is, whites who are poor behave similarly. Thus, the authors are able to create a number of testable hypotheses. Read the paper for more details.

Alex seems to contradict himself here.

From the top of the post

We demonstrate that these differences exist among virtually all sub-populations, that they are relatively constant over time, and that they are economically large.

From the comment at 3:19 PM

What they find is that relative income is doing most of the work. That is, whites who are poor behave similarly.

I interpret the quote at the top as the opposite of Alex's statement.
So, which is it?

Brad, there is no contradiction but I take your meaning. In the second statement I should perhaps have written "whites who are poor with the same reference group behave similarly" although more accurate this is a little mysterious but remember that any status theory requires an answer to the question status with respect to what group? See the paper for more details on how they define groups.

My main point with respect to Robert et al. is that the authors argue that situation not preferences explains most of the difference in conspicuous consumption.

I'm with robertdfeinman on this one. Yet another case of white being defined as the norm and non-whites as the deviant "other". Much like the definition of women with respect to male norms. Never mind that they use the signalling model to show that it's situation not inherent/cultural/genetic/whatever "bad behavior" that matters. ) Leaving out housing is a fatal flaw as this is probably even more a display good than clothes for many. Notice on p17 that blacks and Hispanics spend LESS on home furnishings than their white counterparts, which I assume means controlling for everything else like income.

So all that is going on here is that whites and blacks use different goods to display position, which seems to be entirely driven by their peer group/ reference group's behavior. Whoop-di-doo! This is hardly a blinding insight. I started buying (more expensive) European-made clothes when I moved to Europe, too: my old wardrobe looked shabby compared to the norm here. What is more revealing is that Mr Tabarrok thinks that this is "politically incorrect". The conclusion isn't, especially: situation matters, not race, as Mr Tabarrok finally makes clear. But it pushed all those American buttons about blacks, didn't it?

I don't know what depresses me more: that someone decided this research was important enough to do in preference to some other use of their research time; or that it probably will get into the QJE (note the non-gatekept version has "QJE submission" in the filename and two of the authors are from UChicago. And they use recent US data, which as we all know is representative of the whole world for all time*.)

* Yes, I'm being sarcastic. A plausible interpretation of the data is that the snappier dressing of black workers in white-collar jobs compared with whites in the same job is simply because whites got slobbier in the 1970s and later, while blacks continued with the suit, tie and hat norms of the 1950s and 1960s.

"People without adequate income won't spend as much money on things with no proven benefit."

Ah, yes. That's why the lottery is so unpopular with the poor.

First thanks for proving my point about prejudice disguised as Puritanism.

You might want to read the classic book on the inner city mentality:
"Tally's Corner: A Study of Negro Streetcorner Men (Legacies of Social Thought)" Charles Lemert

But, don't let the data interfere with your thought processes...

Whites are the norm because they are the majority. If were comparing the behavior traits of different kinds of chickens, the norm would be the majority group.

I wonder if any of this is due to inner-city poor people not having access to discount stores, outlet malls, etc and paying higher prices on the same goods?

You should visit the inner-city more often. The inner-city neighborhoods are replete with such places. K-Momo is a popular store here in the Phoenix metro area, and they are – if you ask a young black or Hispanic person hereabouts – the bomb because they have the best hip-hop styles at prices they can live with.

We also have a couple of Nordstrom’s outlets where white customers are not the majority. The Dilliard’s outlet closed up, but for years it sat right on the edge of the barrio, and again, white folks were the clear minority of customers.

Nathaniel said: "McMansions are appreciating assets"

Uh, not any more they aren't. And the display characteristics (granite benchtops and twin washbasins etc) never returned more to the value of the home than they cost. The appreciation is all in the land value. Or don't Americans know about buying the worst house in the best street?

Ray G: "Whites are the norm because they are the majority."
This doesn't explain why males are also treated as the norm, which I have already mentioned. You really don't see what it is you are doing, do you? By defining one group as the norm, their behaviour is automatically, often unconsciously, defined as right and proper. The behaviour of the "other" is by definition deviant and bad. Exactly as the authors have done by excluding home furnishings. One could just as easily write papers about: why white Americans become slobby dressers compared with other groups; or why high-income whites take drugs and drink in ways that high-income blacks apparently don't; or why whites are so afraid of difference that they move to the suburbs rather than stay in racially mixed communities. But those papers don't get written. Why is that?

(Did you know that in the UK it's the whites that perform worst at school, once income is controlled for?)

Robertdfeinman: "I don't see any sign of this paper having been vetted by anonymous, disinterested referees. "
Of course it hasn't. It's an NBER paper. Because economics has made such a mess of the publication process, with enormous publication lags and interminable revision rounds, the "working paper" has become the currency of choice. If you are a member of NBER (i.e. in America and one of the in-crowd), your work will get noticed and accepted even if it's wrong.

Anyway, all peer review will do is make them add a bunch of footnotes saying, yes, they tested these alternative functional forms, and then completely rewrite the paper a few times so that the reviewer can follow it while sitting on the bus giving it much less than full attention.

American white males currently dress with as little flair as any time I'm aware of. The black-white peacock gap is a lot wider today than in, say, the 1970s. I think the Silicon Valley (non-)style has had a lot of influence on how white men dress going back to the early 1980s (my old Chicago marketing research firm adopted a "business casual" dress code specifically so we would look "high tech" to investors when we went public in 1983).

Research has shown that a beauty premium exists in wages, so defining money spent on making yourself more attractive as conspicuous consumption or status seeking is in its self questionable. The Black white differential in spending could just as well be explained by a larger premium for Blacks.

Since the US savings rate has declined steadily as the top 20% has increased their income share, the comsumption habits of this group would be of more interest. The assumption by economists about this group's marginal propensity to save seem to no longer hold. It is a puzzle that has real policy implications.

Ugh. Steve Sailer and his troupe of biological determinists have arrived, which means this thread is officially dead.

If it's rational (and a-ok) for car salespeople to charge more to black customers, why don't house / home furnishing salespeople charge whites more, since they are the goods white buy for display/status purposes? (Refer p17 of the paper that started this off.) One could easily imagine offering bigger discounts off the advertised price to black customers than white. If that doesn't happen, maybe we are back to having to admit that some people are prejudiced and discriminate, instead of telling dubious genetic just-so stories about white teeth.

My next QJE submission, "What genetic and evolutionary factors have caused white American males to be so badly dressed?" The policy implications of eliminating the associated negative externality could be profound - we won't have to look at them and they won't wonder why their wives lose interest after a few years.

Im not sure if this is a typo:

All Total White Asian Black
Gifts of goods&services:
Education ............. 200 221 215 388 45

(Last page,second last item)

This cant be correct?

The variance is just too high!

Suggesting racism might play a role in this discussion is preposterous .
By the way, are you Jewish?

Mr. Feinman:

A better example of Saileresque debate could not be found. Snarky ad hominem, a dodge disguised as an attempt to educate and then a call to "reality".

Do you have ANY evidence or data to back up your claim?

why don't house / home furnishing salespeople charge whites more, since they are the goods white buy for display/status purposes?

Unfortunately cars are one of the only products still sold using bargaining. But pick any number of home goods stores. There's nothing that can't be found at a discount store like Wal-Mart for orders of magnitude lower prices. I know of an oven importer who has all their ovens made in China. Some boxes say Made in China, some say Made in Italy. Guess which one costs 30% more? Ever heard of Whole Foods Market?

Thoughts on shoes:

1. Children under 18/household = 0.6 for whites, 0.7 for asians, 0.9 for blacks. Kids outgrow shoes quickly, no matter how xheaply you buy them.

2. Cars/household = 2.1 for whites, 1.6 for asians, 1.3 for blacks. Blacks have to do more walking than everybody else? Just a guess, since they also spent less than everybody else on public transportation [tenuous, I know].

3. Sex of reference person/household [male/female] = whites 52/48, asians 62/38, blacks 36/64. Black heads of households, and therefore probably the primary earners, would appear to be women. In office settings, men can usually wear only a few different suits and a couple of good pairs of shoes and still look "professional;" women are generally expected to have more and noticeably different outfits than men -- including coordinating shoes -- to look "professional." It's unfair, it's sexist, and it's old-fashioned, but it's one of those unwritten dress codes that you have to watch out for in the workplace.

4. Black households spend only about 150-160% more on footwear than the others. At least some of that difference could well be for strictly utilitarian reasons, rather than signaling anything. I'm not sure it's especially valid to be using footwear expenditures to buttress the signaling and bling bling arguments in the NBER paper.

Speaking of the NBER paper...

1. Looks to me like some fuzzy thinking went into delineating "visible consumption."

2. The race categories are divided into black, white, and hispanic, but somewhere in the appendix is a note that all mixed-race heads of household were assigned to the hispanic category, because leaving them out of the analysis didn't affect the results. If leaving out mixed-race households doesn't affect the results, why not just leave those households out of the calculations entirely?

3. Somebody else mentioned the discrepancy in reporting averages for some of the results and transformations of the data as other results. I did a bit of quick-and-dirty ciphering, mostly just on some of the averages, and couldn't quite get everything to add up. A little bit of handwaving going on?

I dunno, the slightly fuzzy definiton of visible, the slightly cavalier treatment of race, and the slightly obfuscatory math [not to mention the U of Chicago connection], all raise red flags for me. I'd want to see the raw data on this one.

Hard to keep track of all the generalizations here about groups of tens to hundreds of millions of people based just on Table 2100. I started to make a list but you know it was just too long for an appropriate post.

I do have to say this was IMO the looniest remark of the thread:
"Merely being white is a way of signaling wealth. If you don't believe it, visit Peru."

Sorry, when I was growing up in Eastern Kentucky, I missed that class trip. All the white folk in my town must have been signal-blind 'cause wealth didn't get signaled too often back then.

Yep, its white peoples fault.

Thats why they all put rings around their necks and gigantic metal discs in their lower lips in Africa too, cause somehow, whitey will judge them if they don't.

... Getting tired of this.

hi,I University majoring in the legal profession.After graduation,I 徵信 the work of the strong interest.Has worked in several徵信社.Has a wealth of experience. Now I immigrants France,Hope to continue to engage in the work of徵信 credit.
now,is to wake up every day to drink å’–å•¡, shopping. I hope that early awareness of Boles.
thanks,thank very much.

Comments for this post are closed