« More Thoughts On Selling Cars | Main | Enron »



speaking of shoes :

the Consumer Expenditure Survey


says blacks spend more on their shoes than whites

in 2003
Expenditures on footwear by whites and other races: $274
Expenditures on footwear by blacks: $440.


You write about visiting Texas and Oklahoma as though they were in a foreign country, where you wandered around like some guilty liberal, no less, hoping not to tread on local customs. Next time you visit, relax a little!

Truth be told, here in Texas you can talk to women, actors, graphic designers, and yes, even black people about college football, too. (Ask any black 12-year old boy in Texas about Vince Young in the Rose Bowl to see what I mean.) That said, most of us are smart enough to also talk to you about any other subject you might normally address with your colleagues in New York.

Self-examination of one's motives and actions is useful and important, I suppose, but sometimes a cigar is just a cigar, and I think you make too much of the "success" of this conversational tactic. It sounds like your assumptions about the people you met in Texas and Oklahoma may have done more to frame your experiences with them than anything real about them as individuals.


Malcolm, you are good. I'm gonna' keep my eye on you.


its almost a new year, people. can we talk about something else for a while? please?

Fernando Valério

Dear Malcolm, I am from Brazil, I'm a doctor and my skin's color is white. So, my life is very easy, because everybody treats me very well, think I have a lot of money.
But, for black people, beside my country has a lot of Africans sons and our culture be very rich in africans behaviors, the education and health are very poor for blacks. The number of deaths in blacks are very higher than for whites.
Today, our president reserved some places in the Universitys for blacks. This atitude seems to be very nice, but only incresead the racism.Fernando Valério

Sue Denhymn

I missed the issue of the New Yorker with your article on predictive models that determine successful movies. (Was your name on the cover? I only bother to read those with your name on the cover.) I was looking for it online when I found your blog.

I build predictive models in my day job, mostly decision trees and logistic regression, but not much neural nets because they tend to overfit. In marketing, one uses anything that may increase ROI. Giving higher starting prices to black men may, on AVERAGE, lead to higher sales. No racism (or any ism). Until the word gets out. That’s where the software stops and the “art” begins.

I read your definition of racism. It’s too compact. Racist rants like Mel’s are clearly from deep inside him. Michael Richards was angry and he lashed out to hurt. I’ve been called a bitch by people who don’t hate all women, but only want to hurt me. You are a scrawny, mulato, Canadian (spit, spit, spit). Only the last was a slur, and it was given in jest (sort of) and none were meant to hurt you.

Two dogs (perritas) are driving in Texas, they have a flat tire. One goes for help, returns and says, “Nadie no quiere prestarme un gato.”

Sydney Mogapi

Hi malcolm,

I am a young black professional (Economist)from South Africa. I stumbled upon your blog by chance and I was deeply touched by your ability to succintly elaborate a point. I really think your a great and some people in this blog can learn a lot from analyzing your comments without prejudice or filters. In South Africa we deal with racism a lot, as you can recall, it is the a country which took racial discrimination to new level by institutionalizing it and reducing those of a darker hue to being non-persons. We are as a nation trying very hard to move forward and make our nation a rainbow nation as per Mandela's views.It is a difficult process, but I believe it will be done. The first thing that we need to do is confront the fact that even though we have one of the most liberal constitutions in the world, racism or forms of racism exist. A perculiar development which aims to offset our path towards reconciliation is that some individuals have chosen to "mask" incidents of racism as something else which is painted as natural and rational.Just like some bloggers in your website try to to potray the actions of the car salesmen as rational when they are in fact manifestations of racism. I might add that there is nothing rational about the car salesmen's behaviour because by inflating the price by $1000 based on race they may actually increase the probability of losing the sale.That is not rational. That statement about black people wanting to pay more is condenscending and downright racist, in South Africa if you made that statement we will take you to the Human Rights Commission which was specially enacted through our constitution to deal with such issues. Please do not worry about so much opposition to your views as most of those who oppose you are those that want black or any other prejudiced race to think that there is nothing unnatural about the way they are treated so as to perpetuate these inhuman acts. It is also obvious that Sailor is a racist. DO NOT RESPOND TO HIS COMMENTS, HE IS AN IDIOT.

Susan RoAne

I am a great fan and enjoyed meeting you at your booksigning here in Marin County.

The truth is that conversations with strangers are dicey. That you knew how important college football is in certain states is a testimony to your awareness and WILLINGNESS to try a topic.

IF your conversational partners hadn't responded, you would have gone to another topic until you found common ground. And an organic exchange would have occurred. Even I (a Fighting Illini) would have been responded to a college football conversation.

As for the car salesmen issue, I have heard stories of dress/race/ethnicity/gender assumptions/pre-judgements that lose sales. Being open-minded is just smart business. Why? Because 'ya never know'.

Nick Hodges

I don't think the salient piece of information here is what the black men were initially /offered/, I think the salient piece of information is what black men eventually /pay/.

Perhaps car salesmen, on the aggregate, have the perception that black men drive a harder bargain. Perhaps salesmen have perceived -- again, rightly or wrongly -- that black men /will/ pay more for cars.

Is there any evidence that they tactic causes car salesmen actually to lose sales?

Or, put another way, is it reasonable to assume that car salesmen, as a group, are a bunch of racists? Or is it more reasonable that they are acting rationally? What would Occam's Razor say here?

Matt Searles

I keep thinking this has something to do with the "relationship to the symbol." Some of the stuff brought up in Blink was strikingly close to Jung's association tests... the test where you test how easy or hard it is to associate someone who's white or black with a given thing.. I'm a white guy from New England.. but for whatever reason I've become interested in race matters and race politics and african american history.. and stuff like that.. and I find that studying that, effects how I relate to the idea of whiteness and blackness.. I also kinda grew up on Frank Zappa who would always make fun of white music.. so I imagine this effects my relationship to the idea.. in a certain sense..

So what I see you saying is that the idea, perhaps, is not inherently bad.. which has to with identity and relationships to identity, and stuff like that.. I mean.. well you know it doesn't make sense to me.. people who associate asians as being bad drivers or italians with organized crime.. when I think asia I often think Akira Kurrosawa, when I think Italian I often think of Copella's interest in filming people eating around a table.. and a certain family thing.. and something to do with neo realism and Falini... But I know that other people have really dumb ideas about this group or that group, or whatever group... and not just about groups.. about everything under the sun.. I mean.. common wisdom seems like a sorta flawed thing sometimes.. or not that common wisdom is flawed.. but there's a certain kind of something in American culture, at least, I think... where main stream TV can be so shallow.. and it encourages a certain shallowness.. and we are to take our presumptions on the nature of reality more seriously then we do reality.. And of course the idea that our idea of reality might get in the way of our seeing reality.. that's an ancient spiritual issue.. I mean that's what Buddhism is all about trying to help you get over...

I notice when I talk to strangers I often feel like.. I'm some sorta John Von Numan / John Nash sorta game theorist.. in the sense that I drive down different roads.. trying to get a read on certain layers of the persons consiousness.. attitudes about this that or another thing.. and eventually I'll get so I'll feel like I've read the person in a certain way.. and you know.. you start off with the clues you got.. and you play with your own conceptions of probabilistic frameworks and..

I don't know what I'm trying to say.. Maybe it all comes down to.... I need to read more Joyce...


"The short answer to that question, I think, is that this is what racial prejudice is: it is the irrational elevation of race-based considerations over other, equally or more relevant factors"

Gladwell hasn’t provided any evidence that Black professionals are not just as likely to be willing to overpay for cars as Black non-professionals, therefore his conclusion about the actions and motivations of the salespersons is based on an unscientific assumption that car salespersons are racist. Unscientific assumption are generally based on prejudice (pre-judge), given that assumptions of anti-Black prejudice are most usually touted as an indictment of white attitudes, we can conclude that Gladwell bases his interpretation of the behaviour of care salespersons on his racist assumptions about white people.


A very similar observation forms the basis of noted economist Amartya Sen's Identity and Violence - The Illusion of Destiny. Like you, Sen avers that each one of us has multiple identities and associations. But but by giving undue focus on only one of those numerous identities, to the exclusion of all others, - most frequently religion, race, country and color - we create stereotypes. So all Muslims are expected to live and die for Jihad, and all Jews are necessarily money-hungry (notice the almost consistent potrayal of a Jew as the shrewd banker in several works of fiction).

Also, one stereotype leads to creation of several others. So, for instance, a stereotyping of the African car-buyers on the basis of their race by the white salesman, is in turn likely to lead to a stereotyping of 'whites' by the Africans, again based on race. And so we get a vicious cycle, in which we get caught deeper in the maelstorm of the 'single identity association', and it becomes increasingly difficult to break the Chinese wall and establish trust.


I can find only three explanations for the data presented: 1) The practice of discriminatory pricing is inefficient and salesmen are not rational maximizers; 2) the practice is inefficient, but salesmen hate black people so much that they're willing to essentially spend money to rob them; or 3) the practice is efficient.

Explanation #1 poses serious problems for economics in general, so I place it outside the scope of this discussion, if not outside the range of possibilities. Explanation #2 implies that people are willing to spend money on racism. While it is entirely believable that people might invest in racism (e.g., in a situation where doing so would protect their valuable existing prerogatives), I opine that in a large, modern, urban market, no salesman could establish a positive ROI on a personal investment in racism; therefore, I dismiss #2.

I conclude that the practice of discriminatory pricing is efficient, i.e., it results in higher profits over time. Speculation about the social politics of race and wealth provides a simple, if unverifiable, explanation for the dynamic that creates this situation when a black customer negotiates for the purchase of a symbol of social status and wealth from a white salesman. Simply put, buying a car is largely proving something to oneself and the world.

It's not an exact analogy, but consider the dynamic at work when a man sets a price with a prostitute. Can he afford to haggle?

T. David

I agree that charging different prices to customers based on race is immoral (and less importantly, probably illegal). I can understand how annoying it must be to someone who is being charged the higher price. However, such discrimination seems qualitatively different from the sort that used to result in lynching.

I keep thinking about the study from the point of view of the commissioned salesmen, having been one (stereos) many years ago.

There are two components to making money at commissioned sales: volume and profit. The more a customer actually knows about the product and its pricing, the less likely it is that the ultimate sale will have a high profit, that is, a high price for a particular product. (Perhaps it is also immoral to charge more to the uninformed, but that is a different issue.) At the same time, the more money a potential buyer has, the more likely he will ultimately buy from someone. At the early stage of negotiations, salesmen have little information on either subject.

One thing the salesman does know is that the customer is never going to buy at a price higher that the initial quote. He also knows that a knowledgeable customer may simply walk if the first quote is too high. If the customer also has the ability to buy, the salesman has lost any chance of even a low profit sale, which is better than nothing. Conversely, a customer who is not going to make a purchase should be quoted a high price and the discussion should be ended as quickly as possible. Salesmen need to be good at determining who will and will not actually buy. Inability to correctly identify “strokers,” i.e., non-buyers, is an identifying mark of inexperienced salesmen.

Notable about Ayres’ study is the fact that none of the customers bought a car. From their point of view, the salesmen failed in each case. I suspect that customers who had done their research before visiting the dealerships (reading Consumer Reports, determining dealer cost, etc.) could actually purchase cars for the same price irrespective of their race or sex. I also suspect that the discrimination motivating Ayres’ salesman is their assumption (based on insufficient data) that white males are more likely to have done research and to have the means to complete the transaction, not any underlying animosity toward minorities or a desire to confer benefits on a privileged class.

The mistake Ayres’ salesmen made (from their point of view) was in failing to identify the white male “customers” as non-buyers. Could it be that the salesmen were able to “thin slice” and correctly identify non-buying blacks and women, but could not do so when it came to white men?


I agree, Malcolm missed the mark on this one. It is my responsibility as a seller to maximize my profit, and if one class of people is consistently willing to pay MORE for my product than another, then why should I give money away for the sake of an abstract concept of fairness? After all nobody is putting a gun to the buyer's head and making them fork over the money. I live in a country where I am regularly charged more than the locals, based on the (correct) assumption that I have more money and am less willing to haggle for that last dime. The proper remedy is for blacks to not agree to pay the higher price. Caveat emptor.


Quoting on the basis of race and gender is similar to making random opening quotes, though more socially pernicious. It seems the salespeople are using a fairly crude heuristic to compensate for the universal difficulty of correctly identifying customers' price sensitivity (see The Undercover Economist, by Tim Harford). Both the random approach and the race/gender approach are likely to be more efficient than simply putting fixed price labels on all the products. The ideal salesperson would be able to
elicit relevant information to identify a potential customer's price sensitivity and then give a corresponding opening quote. In practice this is difficult/impossible. Experts tend to be overconfident in their judgements and I suspect that may be what is going on here. Because they can't really have the kind of skill they think good salespeople are supposed to have, they rely on a dodgy heuristic and pretend to themselves that this is skill in action. The delusion can be maintained because the heuristic is more efficient than price tags. I wonder if new car salespeople need to be taught the race/gender heuristic, or if they arrive on the job with it already installed in their brains.

T. David

It is taught, or at least, it used to be when I was in sales, but not by management. Rather, a new salesman will hear comments or be made fun of after he wastes hours trying to sell to someone the other salesmen have identified as a poor prospect based on sex/race/ethnicity.

David Milton

I may be stereotyping myself by saying this... but try telling any black man that he's a 'sucker' to his face. All the black men i know are always trying to hustle their way into a better deal. They are stereotyped as being dumb. Do you think they don't know that many people think that way about them? It's in everyone's instinct to strive to survive. Maybe blacks act the way they do because they feel everyone is trying to take advantage of their lack of intelligence. But whoever put that lack of intelligence idea in their head? What makes me think i know what i'm talking about? I noticed that a few people disagreed with this blog. The reason for this is... everyone wants to be number one... to be the one up man.. better than everyone. Instead we should look at all aspects and not judge, and accept the situation due to the reasons for influence by that individual. You almost have to forget who you are, what you learned, your influences, and power from confidence to see all angles. Most people don't know how to interpret a person's influence for impression because they're too busy judging people based on their own influence. I don't disagree with why anyone disagreed. It's in everyone's instinct to believe that their views on situations are the only ones correct due to what's influenced in life and past experiences. And this all ties in with why a black person was treated a certain way, or why a salesman was influenced to express their impression.

Art Smith


I hope you are still looking at comments on this topic...

Your Texas Oklahoma example served its purpose of a viable reason for sterotyping.

As a white, male, Texas, business man I would like to add a point.

In a situation where you are communicating with someone directly why make such crass assumptions about them. Why not just ask about the weather and then let the conversation take its own course?

For the record, even though I'm a stupid, unedumacated, truck drivin, boot wearin, beer guzzlin, gun totin, dumb a**, arrogant, red neck, hick from Texas I can discuss topics other than football. Some examples: the wonderfulness of W as President, or why damn yankees pee me off, or how come people from England can't pour pee out a boot with the instructions on the heel. Or, maybe we could talk about our favorite cuisine. I prefer Japanese but can't go but a few days without Indian. What if we discussed one of your books. I'm fond of Blink.

Instead starting with sterotyping why not start with "Howdy. What's on your mind today?"

Art Smith
Fredericksburg, Tx


Was wondering what anyone thought about NBA ref's racial bias story in NYT 2day, cheers!



Not sure if its relevant at this point but why not?

All people make judgements based upon their own experience set/world view. No matter what anybody says, being white or black carries with it a certain subcultural burden. We all get raised and recieve our initial information set. We live, and every day our information set grows and changes. Every identity has 3 parts. How we see ourselves, how other people see us, and how we think other people see us.

Being white and from rural america, I did not have a lot of experiences with my black friends. So the black people I saw were judged against the face black america projects of itself, which is unfortunately less than flattering. Does the face actually represent the totality of the black community. No. But with no other information to make judgements on. Joining the army, and being the rice krispy in a box of cocoa puffs was an eye opener. The dataset grew.

Black people are equally lacking of information as well. So they make their judgements off of an equally skewed dataset. The very real consequence of "white flight" is that minorities dont really have a full picture of white america, and white america is still stuck on the plummeting property values and school quality, and increasing crime rates which are more very real unpleasant things that correllate.

Change the datasets, and bring the 3 selves into harmony in both communities, and maybe someone would have something more productive to do that glare at each other.

Seriously, everybody is racist. The real enlightenment is understanding that and moving on.


The stereotying isn't lousy, from an amoral profit-maximizing perspective, if it works. Perhaps upper-income blacks are even more likely to negotiate ineffectively because they either aren't good at it or don't care


If some form of clothing were associated with hard bargainers, and whites wore the clothing of the weak willed and the blacks of the strong negotiators, the black-white difference would probably be smaller.


Juno makes a good point... the most important factor in trying to make a business deal is the mental disposition (attitudes, beliefs, etc) of the buyer. It doesn't matter what you as a seller offers; if the buyer doesn't want to pay that much, then the case is closed. A buyer's height, weight, education, race, age or anything else doesn't matter as long as his mindset is that he doesn't want to pay a certain price. So the process of selecting what price to offer a buyer is, ultimately, a guess as to what the mental desposition of the buyer is.

So the seller has to guess about the buyer's mental disposition. What the Ayer's study shows is that sellers use race and gender to make predictions about mental disposition, even in the face of other data such as education, socioeconomic status, and occupation.

I think Gladwell's point is to put it mildly, strange to make a prediction about a person's mental disposition based on race when you have other information that could probably give you a better guess.

It's a game of sellers guessing how buyers think. What gives you a better guess about how a person thinks, his race or his occupation? 'Black' may suggest 'little knowldge and not good negotiating skills' and 'bank executive' (which I think is the occupation they tended to use in this study, although I could be misremembering) should suggest 'much knowledge and good negotiating skills.' When confronted with these contradictory pieces suggestions, which one should be the safer bet, race or occupation? 'Black' also gives suggestions opposite of those suggested by the other factors in the study, such as socioeconomic status (all of the buyers were said to be from a richer part of town) and education.

The best business strategy is the one that allows the seller to get the best guess of the buyer's mental disposition. I think Gladwell's point is that when trying to make the best guess for how a buyer is thinking, why would you count race at all when all the other information you have suggests the opposite of what race suggests? Race is only one of many clues you use to help you make a guess, and if race suggests a person might have one disposition but all of the other evidence suggests that a person will have a different mental disposition, then race probably shouldn't be counted as evidence.

It's kinda like weather. If it's December, then the fact that it's December is a clue that you can use to help you guess what kind of clothing you should wear. But if it's 80 degrees outside, the information you get based on the fact that it's December should be ignored in making your decision. Saying 'He's a well educated upper class bank executive, but he's black, so I'll charge him a higher price' is like saying 'It's 80 degrees outside, but it's December, so I'll bundle up and wear my heavy winter coat and gloves.'

The fact that it's December is usually a good way to predict what clothing you should, but if there is enough evidence predicting something else, then you should disregard the fact that it is December in making your final judgement. In the same way, a buyer's race might be a good way to predict his attitude when he goes to buy a car, but if there is enough evidence supporting a different prediction, then you should set aside his race in making your prediction of his personality, preferences, skills, and etc.

The comments to this entry are closed.

My Photo


  • I'm a writer for the New Yorker magazine, and the author of four books, "The Tipping Point: How Little Things Make a Big Difference", "Blink: The Power of Thinking Without Thinking" and "Outliers: The Story of Success." My latest book, "What the Dog Saw" is a compilation of stories published in The New Yorker. I was born in England, and raised in southwestern Ontario in Canada. Now I live in New York City.

    My great claim to fame is that I'm from the town where they invented the BlackBerry. My family also believes (with some justification) that we are distantly related to Colin Powell. I invite you to look closely at the photograph above and draw your own conclusions.

My Website


  • What the Dog Saw

    buy from amazon


    buy from amazon

    buy from amazon UK


    buy from amazon

    buy from amazon UK

    Tipping Point

    buy from amazon

Recent Articles

Blog powered by Typepad