Jamelle is the second person to say this to me, in re my credit check at the AT&T store:
Congratulations Megan McArdle, you just had your first taste of what being a black person is like.In all seriousness though, African-Americans are far more likely to be subject to those kind of credit investigations than a white person of equal means, since on average, African-Americans are more likely to have some sort of serious credit liability.
As far as I know, they run credit checks for any monthly cell-phone contract; that's why people with bad credit have to get pre-paid phones. Certainly, they checked my credit when I bought a broadband modem, which starts at $40 a month.
Credit is one of those weird areas where there is a lot of belief in discrimination, but as far as I can tell, not all that much evidence. Most credit checks are part of an automated procedure that either happens or it doesn't, and most loan issuance is done virtually automatically by a computer that either says yes or no based on your credit history. Now, there are border cases that require human review, and it's possible that had my name been "Malika" instead of Megan, they would have turned me down.
But if that were happening on a widespread basis, loans to minorities would be insanely profitable. They're not; rather, they seem to be about as profitable as other types of loans. Yes, I've seen the research arguing that people in black communities get worse loan terms than their credit score suggests. As far as I can tell, this research failed to control for some pretty major factors, like assets.
I don't want to go to far down the Gary Becker line--it's possible that companies could all be behaving irrationally. But the evidence that they are seems pretty thin--in fact, just barely solid enough for plaintiff's lawyers and journalists to revive it every few years. If the companies were statistically discriminating against African Americans, giving them worse loan terms than they really qualify for, they should be paying off those loans at higher rates than whites.
They're not. Most of the aggregate research I've seen fails to reject the null hypothesis that there is no discrimination in loan markets, which means that if there is discrimination, it is not catching huge numbers of people who are more likely than their loan terms would suggest to pay their bills on time. Just to be clear, we're not talking about research that says that blacks who get a higher interest rate don't pay off at the same rate as whites who get a lower one--you can't blame the default rate on the higher interest rate. We're talking about the fact that minorities do not outperform their own loan class. If loan companies really were discriminating, issuing subprime mortgages and car loans to credit-worthy minorities should be a license to print money.
The evidence for discrimination in the labor market seems strong--nay, nearly incontrovertible--to me, at least at lower skill levels. And it's clear to me that African Americans have a lot of structural barriers to wealth accumulation, But I remain unconvinced that credit rationing is one of those barriers.
But Malika is such a beautiful name...
As far as I know, they run credit checks for any monthly cell-phone contract; that's why people with bad credit have to get pre-paid phones.
Madness. No one ever checked my credit for getting a phone. This is new and crazy in most contexts.
it is worth noting that "Malika" (a reference that has obviously flown several thousand miles above my head) is only one "l" from "Mallika" (MULL-ee-ka)which is not an uncommon girl's name in north and south india.
I have also noticed "Ananda" floating in the same Desi-Black naming nexus.
themightypuck-
They did run a credit check for your cell phone contract.
You gave them your social security number and signed a long contract that had a paragraph allowing them to check your credit.
Your credit gets checked more often than you realize.
Not to hijack, but I believe Malika, as a popular name among black folks, is directly related to the relatively large number of African-Americans who converted to Islam.
While you're right that claims like this require evidence, it's also true that acquiring that kind of evidence is going to be pretty difficult, considering what lengths these companies would go to in order to hide that info. But you're right, not enough evidence at present. I just don't buy that it would be irrational for companies to do this because it would reduce the efficiency of their market; when they exhibit bias in their hiring practices
, they are reducing efficiency by potentially eliminating otherwise qualified applicants. Markets are efficient; corporations are not.
Plus, in a bit you'll have a bunch of people logging on to tell you how it would be perfectly rational and in fact responsible for these corporations to discriminate against black people, and how it's just your damn liberal guilt that prevents you from saying so.... And if they exist out there to comment on here, there's probably some who think like them who work in credit. Not proof positive, of course, just something to think about.
Thanks for the link Megan. And for the record, the post wasn't meant to be a dig at you, just a way of launching into some stuff about credit and African-Americans.
t Jayrwasdff,
I didn't read the contract so it is possible. I suppose it makes sense since providers are really selling phones on an installment plan.
I know nothing of the cell phone business but it seems to me the marginal cost of a phone is pretty low. It makes no sense to me to discriminate unless your marginal service costs are high or you feel people might be gaming the system. I can see this happening with iPhones where people might buy, unlock, resell, and break their contract. For most other phones it seems nuts. Depending on how much the contract is subsidizing the real cost of the phone, it seems crazy to lop off potential customers. I'd be curious what the risk calculus is for providing service to people with even abysmal credit.
Jamelle--I didn't take it as a dig; it just struck me that this is a widespread perception, with surprisingly little empirical basis. (I wonder, too, if self-segregation doesn't have something to do with it--if all your friends are black, it's probably easy to assume that they're checking your credit *because* you're black).
Freddie--it woudn't be irrational to charge blacks higher prices if you could get away with it; rather, it would be irrational not to set up a company that charges them slighly lower rates, taking all their business . . . and for another guy to do the same to you . . . until the arbitrage opportunity disappears. It's not that I don't believe discrimination exists, but it's harder to maintain in situations where it's very costly, like competitive loan markets. I would expect that if we didn't have a highly competitive securitized loan market, you'd see at least some detectable discrimination today.
But we do rely mainly on computers, who really "can't see" race, except through blunt proxies like zip code. Given the data on loan profitability, I would say that this particular form of discrimination is presumptatively not significant unless someone comes up with a better test.
Good point about the computers. I guess this means that when our robot overlords rise up to conquer us all, we'll finally have racial harmony.
It's gonna be great. I can't wait to fight in the resistance. Robots are right up there with zombies and Nazis when it comes to foes you can have absolutely no bad feelings about killing.
Without commenting on the argument for 'rationality', the theory simply does not conform to reality. The market does _not_ eliminate discrimination, or indeed, do much to ameliorate it. But point this out, and the usual apologetics follow.
It's like what has happened in my town: for years the black community has claimed that the police harrass them, and have trotted out statistic after statistic. To which the police give the ritualistic reply that this isn't true, that they don't condemn racism (that's not what the black cops say!) etc. Finally a comprehensive study was done, everything from following through on calls for police assistance to investigating complaints of police behaviour to random traffic stops and and ticketing, etc. Guess what? By every single measure, blacks were discriminated against. Police were less likely to appear when requested for assistance. Complaints against police misbehaviour were found to be valid at something like eight times the rate for whites as for blacks, and blacks were over three times as likely to report it. Big surprise - the complaints were found to be most valid when it was a white person complaining about a black officer, and the least valid when it was a black person complaining about a white officer. When the search for a new police chief commenced, an Asian-American found himself frozen out, though he had the highest ratings in a number of categories. Subsequent investigation found emails sent from the former Chief's office computer saying that some way had to be found to get rid this candidate, because the men 'wouldn't accept the authority of an Asian' (though he didn't use the term 'asian.)
And so on and so forth. What was the response? You guessed it: the police department said that it showed that blacks were more likely to complain frivolously, and more likely to be found in violation of the law when stopped . . . and they supported this with their ticketing and arrest records!?!?!?!? The Asian-American? The ability to be an authority figure was said to be a 'valid concern'. This after weeks of denials that any such reasons for discrimination existed.
So, to bring this back to the current discussion, what would be evidence that the market does not punish discrimination? Besides the stuff that already exists that is, and is not accepted by so-called free-marketers; we already know that blacks in a variety of professions have lower incomes:
ScentOfViolets:
As far as I know the situation is a bit more complicated than "markets eradicate racism" or "racism will tend to prevail over markets". I think Tim Harford wrote a fair amount about this in The Logic of Life (which is I think a meatier book compared to many pop economics books).
And Megan explicitly said she was just referring to the credit market, not everything - she actually said the labour market is one where markets have not managed to reduce discrimination. Nor did she even say that the theory of efficient markets is always right - she just said that there is no statistically significant evidence of discrimination in the disbursement of credit.
"The evidence for discrimination in the labor market seems strong--nay, nearly incontrovertible--to me, at least at lower skill levels."
I know this sentence is on the fringe of this posting, but it's prompting a question (which is in part spawned some of the discussion surrounding the "Men at Work" issue in Atlanta.
Since it's apparently a given that women and minorities are paid less than white men for the same work, could someone please explain to me why men who do jobs that can be done equally well by women and minorities aren't systematically replaced by these employees with lower labor costs?
Wouldn't it be irrational for companies not to engage in this type of behavior?
ScentOfViolets:
If you were hoping to prove that markets can be discriminatory, you sort of fell over by using a government department as your example.
ScentOfViolets:
Police is a monopoly, with no competition, and there are no market forces at play. Loan companies are not monopolies, and they have competition. I really don't see how your example applies.
No, I wasn't. I was using it as an example of someone denying that something was so, even though the objective data clearly and undisputably contradicted them. The same is true for the free-marketers and their cure-all tonic. The fact of the matter is that the free-market does not behave the way Megan claims it should behave in correcting inequalities, and there are plenty of studies, thousands of studies, that confirm this. And yet, the free-market ideologues stubbornly refuse to admit they were wrong (or indeed, ever produce evidence for their positions.) Instead, they throw out hypotheses as to why the putative results of the studies are actually in accord with their theories, such as speculating that women get paid less because they have less experience or work fewer hours, or they are differentially concentrated in certain jobs or job positions (that's why I chose this particular study.)[1]
_That's_ why I made the comparison, and that's why I was asking for what sort of evidence would be considered adequate to falsify the hypothesis that the market corrects for discrimination.
[1]Interestingly enough, studies that are much more ambiguous, and which are much weaker in confirming various claims - such as, say, minimum wage studies - are often trumpeted as 'proof' that They Were Right (if you squint your eyes and ignore other studies which conclude that there is no evidence for the claims being made.)
I submit this blog post without comment:
http://www.portfolio.com/views/blogs/odd-numbers/2008/07/09/racism-in-peer-to-peer-lending
As is his usual method, SOV attacks using an inappropriate analogy. The police are a government monopoly- there is no competition to arbitrage the differentials in police service/harassment.
The second example is closer on-point, but note that SOV really didn't understand what Megan was writing in the first place. Here is what she actually wrote:
And SOV replied:
First, McArdle did not write that the free market eliminates discrimination always, only that it is irrational for someone not to arbitrage away the potential profit by not discriminating against people. She also only wrote that this arbitrage is more efficient as the competition in the market increases- where there is less competition, you are more likely to detect true, economically irrational, discrimination. Need an example? Oh, I think SOV actually provided one in his comment, as I noted above.
Then, using the more appropriate example of the race/gender differentials in the incomes of OB/GYNs, he claims this as an example where the free market doesn't punish discrimination. However, the example he provides does show that the market has punished those that discriminate- it did so by the fact that the discriminators pay more for OB/GYN services. Using SOV's philosophical understanding of how the market is working in regards to OB/GYNs and discrimination, then to "fix" this market failure would require that the government force women to not discriminate in their choices of OB/GYNs. It would be analogous to requiring me to not discriminate in my choice of where to eat out, or whose CDs to purchase.
The free market does not guarantee that people won't behave irrationally, only that irrational behavior opens up profit potentials that would not otherwise exist. If people don't step in to take advantage of this (such as, for example, purchasing services from a black male OB/GYN for less than you could acquire from a white male), then there is no scope for government intervention to fix the problem.
ScentOfViolets,
Here are your facts:
1) Black people arrested and stopped more frequently
2) Black people make more claims of police abuse and more of those claims are found to be frivolous
3) Black officers most likely to have complaints against them validated when they deal with white people
Now, if you assume that black people are exactly the same as white people in criminal behavior, the likely assumption is that police are racist.
If you drop this assumption given knowledge of the massive disparity in crime rates between races, the conclusion isn't racism by police but that a race that is proportionally more criminal will have more negative interactions with police.
There's plenty of evidence for the latter contention and zero evidence for the former contention.
Re your point about labor markets: none of these studies are adjusting for the fact that the average black person with a given qualification is going to be significantly less intelligent than the average person of any other race with that same qualification. For doctors specifically, check out this table:
http://www.aamc.org/data/facts/2005/mcatgparaceeth.htm
The average black medical school matriculant has a GPA of about 2 standard deviations below that of the average white matriculant and MCAT score of about 1.5 SD below.
You would expect that the same spread exists all throughout the employment process because of (a) the racial IQ gap and (b) the pretty much universal program of affirmative action. Any given qualification only has value in that it signals the IQ of the person possessing the qualification (a Harvard diploma is worth more because Harvard has higher admissions standards and so higher IQ students). If you're the beneficiary of affirmative action, the signal is not comparable and so the expected results should not be expected to be comparable.
@themightypuck
Cell phone companies don't run credit checks because of the value of the phone. They run credit checks because they want to make sure you aren't going to run up a huge phone bill (ie way over your calling plan or lots of international calls) and then not pay the bill.
Sigh. As usual, Yancey, your reading comprehension seems to be immpaired by your parasitic ideology. I did _not_ make an analogy about discrimination. I gave an example that hopefully everyone would agree showed a willful denial of facts on the ground. Then I asked for what would be evidence that this didn't really happen in real life _before_ the data was collected. Finally, _here_ is what Megan herself wrote:
So it seems that you are simply inserting modifiers that don't exist in the text. Modifiers which qualify Megan's staments into "usually's", "oftens" and "mostlys", and modifiers in mine which say that they exist in Megan's.
Finally, I closed with another example of discrimination that's known to exist, but for which there is still a large body of free-market apologists. The funny thing is, people like Ken and Josh insisted on misinterpreting what I said abou the police, while mysteriously missing the fact that the second example is _exactly_ what they were looking for. Odd that.
Get that parasite out of your head, if nothing else for improving your ability to comprehend what people actually write.
SoV, you're refuting a claim that I didn't make. I didn't say that markets weren't irrational in this way; I just said that if they did this, they would be irrational. In this particular market, the evidence of discrimination is weak; in other markets, like the labor market, it's much stronger.
Uh, white complaints against a black cops mistreatiment are more often validated than white complaints against a white cop, which are in turn validated more often than complaint registered against a black man by black cop, followed last by black complaint/white cop, and this is because 'blacks commit more crimes'? Your logic simply does not follow. The same thing with convictions and sentencing: blacks arrested for the same offense as whites, and with the same criminal record are more likely to be convicted, and blacks recieve harsher sentencing than whites for the same convictions, all other things being equal. But it's all because blacks are 'more criminal', right? Your conclusions simply do not follow, even granting that blacks are 'more criminal'. And I don't grant that at all - like with the police chief, who's 'evidence' was that they were stopped more frequently and arrested more frequently.
Next you'll be telling me that the Asian fellow really shouldn't have been hired, and for what were quite valid reasons. The fact that the chief denied these were the reasons until the emails were produced is just evidence that he didn't want to hurt the poor guys feelings, right?
Cite?
But we do rely mainly on computers, who really "can't see" race, except through blunt proxies like zip code. Given the data on loan profitability, I would say that this particular form of discrimination is presumptatively not significant unless someone comes up with a better test.
Posted by Megan McArdle | July 12, 2008 11:12 AM
Megan,
try this: http://www.targetmarketingmag.com/story/story.bsp?sid=110152&var=story
things, some, have changed since the early-90's
But if this is true, then your remarks have exactly zero content. I could just as easily say that rational actors would each send me a dollar every month, and if they don't do so, why, it just goest to show that they are irrational. Iow, you are trying to make an argument by definition, a definition, incidentally, which goes against another claim about utilility maximization that you've made before. To whit, that people are always and everywhere utility maximizers. Put the two together and you have people that are not acting 'rationally' per your one definition, yet are still, per another, 'maximizing utilitiy'.
This doesn't make a lot of sense.
And . . . you didn't anser the question: what evidence would satisfy you that there is discrimination in the credit markets? Discrimination based upon race, sex, age, or some other nonrelevant characteristic?
And you didn't answer mine, for a cite supporting your assertion.
Realized your ambition to emulate a planarian parasite yet?
Chuckle. I couldn't have done a better job of marginalizing you myself. But then, that's often the way of it, isn't it?
SoV,
Nice job ignoring the evidence re: labor markets and specifically doctors.
On to what you did respond to. You're right, I wasn't explicit. My hypothesis, specifically stated to explain that evidence is as follows:
1) Black officers are more likely to be violently impulsive (than white officers)
2) Black officers are almost certainly lower IQ (than white officers)
3) Officer IQ is likely inversely correlated with civilian complaints (controlled for race)
4) Black people who complain about the police are more likely to do so (than white people) specifically to deflect attention from their own criminal activities
My evidence for the first point is that black people in general are more violently impulsive than people of other races. This is reflected in violent crime statistics. For more details see:
http://mensnewsdaily.com/blog/stix/2005/10/color-of-crime.html
My evidence for the second point is pretty much every psychometric study ever done. The 15 point IQ gap between white people and black people ensures that in any randomly selected group of people that is mixed, white people will have higher IQs. In addition, affirmative action guarantees that black officers are lower IQ than white officers, as they are allowed to join the police force with lower test scores (in jurisdictions large enough to administer tests).
The third point is informed speculation based on the fact that IQ is positively correlated with work output in almost every job, things like credit ratings, and inversely correlated with criminal behavior. Officer misconduct leading to a complaint is analogous to criminal conduct and is also likely to be negatively correlated with IQ.
As far as the fourth point goes, why not simply complain about police misconduct after you've been caught? There's a slim chance that you'll make a bundle of money if someone finds your story credible. People who are more likely to be criminal in the first place will have fewer scruples that prevent them from making false claims.
Your hypothesis, on the other hand, is that the review board is racist against black people, including black officers. Where is your evidence for this proposition? If the only evidence is the difference found in validity of complaints, you've got no counterargument to my points.
As far as the "differences in sentencing and prosecutions account for the differences in imprisonment and conviction rates" argument, this argument fails because the national crime victimization survey matches the arrest data.
This article discusses this in more depth and goes into the (lack of) evidence for race based sentencing disparity:
http://www.city-journal.org/2008/18_2_criminal_justice_system.html
As far as that one police chief goes, I'm not getting involved in a debate about what one person was thinking when he wrote one particular thing without knowing more about the situation and having it be a bit more important in general. I'm sure that if I were one of the involved parties I'd find it important.
SoV:
Has Megan actually said that people always try to maximise utility? I would have thought that there's enough evidence from behavioural economics that people are not always rational - really, I find it hard to believe that there's anyone outside an ivory tower who believes that markets are always efficient.
SoV:
"And . . . you didn't anser the question: what evidence would satisfy you that there is discrimination in the credit markets? Discrimination based upon race, sex, age, or some other nonrelevant characteristic?"
This is an easy one to answer. If you studied default rates by race and found that the group of people given the same rate had (statistically significant) different default rates by race, sex, age, etc. The group with the lower default rate would be the group which is being discriminated against.
A group with a lower default rate would have some members who were rejected for credit when they should not have been. If the decision making algorithm were correctly created, there should be no predictable differences in default rates between any two identifiable groups. If there are, you add that factor to the decision making process until the default rates are even.
Can I point out that, while not logically compelling, most of us live our lives according to anecdotal evidence? You don't know that you don't like the food at a restaurant because someone did a study; you know cause you've been several times and didn't like what you had to eat. This is a common blogosphere tactic where people have different standards of evidence for things that satisfy their ideology. Personally, if many black people say that they have experienced unusual difficulty in getting credit, then I'm inclined to believe them. It's not scientifically dispositive; but it is worth talking about and considering.
No, SOV, I read exactly what you and she wrote. Here, again, is your own writing:
I am inserting no modifiers into what you and she wrote. You wrote, as a counter to Megan's comment, that the "market does not eliminate discrimination....or do much to ameliorate it". She never claimed it will in every case conceivable, only that it is irrational for market participants to not take advantage of this.
Do you think almost all people behave irrationally? If not, then the rest of her argument follows logically- the more competitive the market, the more the irrational discrimination is arbitraged away. Apparently, my "parasitic ideology" is belief that most people will act in their best interests, and take advantage of the pricing mistakes of others when free to do so. I do not expect all irrational behavior, or its effects to disappear.
When the irrational discrimination is based in the law, then I have no problems attacking it-as citizens, I believe in the principle of equality under the law. However, private transactions are different. If there is discrimination practiced, and it does not involve government coercion to enforce or extend, then the free actions of people is the only solution my principles allow. This is where you and I differ, as can be seen in every exchange we have had. It is even a major difference I have with the hostess of this blog, whose position is closer to yours in this regard, than to mine.
SoV, so you don't have a cite, then. You were merely talking out of your ass, or to be more appropriate in your case, I should say, "arse."
Considering you generally want cites, the least you could do is provide one after making such a sweeping, inflammatory, and self-serving characterization vis a vis race and criminal justice.
I'll accept one excuse: that you were working on your dissertation. But then, that's clearly not the case, is it?
Freddie:
"Can I point out that, while not logically compelling, most of us live our lives according to anecdotal evidence? "
If you've got anecdotal evidence and I've got 150 peer reviewed studies, I'm going with the 150 peer reviewed studies. If we've both got nothing but anecdotal evidence, I'm going with whichever has the more persuasive story.
The possibility of credit markets discriminating by race is not a case where both sides have anecdotal evidence. It's a case where one side has anecdotal evidence and the other has studies and basic economic theory. Any studies that the anecdotal evidence side has are likely to be fatally flawed due to the underlying assumption of zero group differences therefore those studies will be uncontrolled on variables that need to be controlled.
Eh, anyone else want to touch this with a ten-foot pole? Check out the blog link, btw, it's a hoot.
Oh, and btw, as I said, I am only referencing a local phenomena of a denialist as an example of denialism:
There's been a lot of stuff like this that's come out over the years, such as the Asian hiring flap, differential conviction rates and sentencing, etc.
SOV:
If you're going to 'chuckle' about someone writing "Tearing you a new one . . . would be a transcendant joy", you might want to tell us who said it, about whom, and where, since those words were not written anywhere on this comment thread. Otherwise, your chuckling (not to mention the constant sighing) makes you sound like a bearded lunatic chuckling over inaudible jokes at a bus-stop while respectable bus-riders edge away and try not to catch his eye.
Do I sound bitter? I'm sick and tired of SOV hijacking every fourth thread on this site with inane and offensive comments. She needs to get her own site.
Lots of stuff? Should be easy to find a cite then.
(Differential conviction rates, btw, are neither here nor there. That speaks at least as much to the skill and intelligence of the criminal in avoiding conviction than it does to bias.)
My favorite part: "the null hypothesis that there is no discrimination in loan markets." So, as a "rational actor": the NH that there are no sharks in this water. That there are no bullets in the gun pointing at me. That...
Oh, yeah: That there are no institutionalized biases in the economics profession. Almost forgot that one...
Megan says:
Yes, I've seen the research arguing that people in black communities get worse loan terms than their credit score suggests. As far as I can tell, this research failed to control for some pretty major factors, like assets...
Part of the reason the studies don't always reference assets is because black assets, generally, are so low. For most blacks, the value of their assets is not much more than their income, which is to say zero. Because black families usually leave little to their heirs, successive generations of black children start with very little to begin with. So, the assumption that assets are even their to be looked at exposes a bit of class bias here...
See: http://findarticles.com/p/articles/mi_qa5468/is_200701/ai_n21291082/pg_1?tag=artBody;col1
Likewise, a fear of rejection, leads many, but not all blacks to bypass the traditional banking system, when seeking credit, thus increasing chances for rejection because they do not have a sufficient credit history.
I must be black! they checked my credit when I got a cell
I must be Arab (Muslim) !!! I have had to go through screening SEVERAL times in the past few years
funny, I always thought I was a "european-american" christian....
"So, the assumption that assets are even their to be looked at exposes a bit of class bias here...
Likewise, a fear of rejection, leads many, but not all blacks to bypass the traditional banking system, when seeking credit, thus increasing chances for rejection because they do not have a sufficient credit history. "
Bias is not anything that results in a different acceptance rate for people of different groups; bias is something that results in a different default rate between members of the different groups who were given the same interest rate.
In other words, if black people have fewer assets, they're less credit worthy and a loan system that gives them loans at the same proportional frequency as a group that has more assets is biased in favor of black people.
By the same token, a loan system that gives loans to people with no credit history because they are a member of a certain group is biased in favor of that group. Not giving loans to people who have no credit isn't bias because one group has more members that have no credit than another; it's prudent lending.
Megan's argument assumes that black borrowers are unaware that they're being discriminated against, or at any are not factoring feelings of discrimination into their choices: thus, if blacks were getting loans at 7% while whites were getting loans at 6%, the result would be that blacks would feel more pressure to pay off their debt, and would thus choose to pay off their loans quicker. That's the argument, if I understand it correctly.
The problem is that blacks are well aware of the fact that they are being discriminated against - hyper-aware, in fact, since they may see discrimination in particular cases where none exists. Now how would you react if you knew or believed that you had to pay a higher interest rate on your loan than your neighbor. You might decide that prudence demanded that you pay your loan off quicker. But then again, you might also say to yourself that that's what the Man wants you to do - to jump through hoops for being black. So to preserve your self-esteem, you might take the opposite course, and take your time paying off your loan, even though from a purely rational point of view that's the wrong decision.
I expect this is what is happening, and that this 'rebellion' effect confuses attempts to deduce discrimination from payback rates.
lampwick,
I have no idea what, if any, interest rate any of my neighbors pays. Do you? Does anyone? It's not the sort of thing that one typically discusses, among other reasons, because it's boring.
The argument that some don't pay off their debts as a matter of racial solidarity/defiance seems more than a little tortured. And what about rent-to-own, consolidations loans, and lottery tickets? Now there are some cases where not playing strikes back at "the Man."
"I expect this is what is happening, and that this 'rebellion' effect confuses attempts to deduce discrimination from payback rates."
1) You've got 0 evidence for this.
2) Ever heard of Occam's razor? Black people have lower IQs, IQ is correlated with responsible behavior like not borrowing money you can't afford to pay back. The most likely reason for black people having bad credit is therefore, ____. Good luck filling in that blank.
3) It doesn't matter to a lender why someone doesn't pay or pays late; it only matters that they do. Lenders don't want to make a profit or hear a really good story, they want to make a profit.
4) Why wouldn't this apply to anyone who gets hit with a higher rate? Anti-lender sentiment isn't exactly strictly confined to the black community.
5) In a world where this explanation holds true any lender who gave out loans to black people at lower rates would have lower rates of default (and force the higher rate lenders to lower rates to compete).
Seriously, there's no ground here to stand on. Stick to sites that censor comments (or college campuses) if you want to make arguments like this.
You PAID for a broadband modem? Unless it's one of those new-fangled wireless ones, I think you were taken. And $40.00/month? For how long?
'Efficient markets' and utility maximization are different concepts. In one sense, the idea of utility maximization is a useful one: if someone is engaged in behaviour that seems to be at odds with the maximization of their personal utility, you might want to consider other reasons for the apparent irrationality (this came up, I believe, in Megan's revealed preferences postings.) For example, one might conclude on the basis of a crude idea of maximizing utility that people would tip more often and more heavily in restaurants they frequented to ensure good service, and much less often and less heavily when just passing through a whistle-stop far from home. This turns out not to be the case, generally speaking. But the conclusion is not that people are behaving irrationally. The conclusion is that people get some other benefit from tipping, the reinforcement of the idea that they're 'good people', or 'civilized', or being gracious. Not an implausible idea. The obverse of this useful notion is it's vacuity; if literally everything is done with an eye towards maximizing utility, then every act is an act of rational self-interest, whether it's robbing a bank or giving a stranger a lift. Iow, a useful operational opproach to research, but hardly an axiom of personal conduct.
Which takes me back to the specific - Megan claims that people and institutions that do not behave in a way that will eventually erase these disparaties in service are behaving 'irrationally'. Are they? Or is there a good, selfish reason for this behaviour that hasn't elucidated quite yet? In any event, all she has is a theory with no data. And since there are many, many, many instances where 'theory' proclaimed that discrimination would go away and that it was 'irrational', yet the theory was completely, gloriously, astoundingly wrong, the default isn't that others have to prove the theory wrong (this wouldn't be true in any case, as a matter of good science), but that she has to prove her theory right.
Is Steve Johnson a Bell Curve enthusiast or something? What other way to explain his apparent obsession with I.Q. as some kind of ultimate human metric? Perhaps IQ does correlate with good credit (does it really?) but wealth would be a more logical (and probably reliable) predictor of credit (someone tell me if I'm wrong here), and wouldn't require an essentially racist analysis of claims of discrimination. And it would just as easily satisfy his cry to not increase entities unnecessarily.
Actually it's rather stronger than that. 'Theory' has suggested time and again that blacks would not be discriminated against economically, justifying this assertion by claiming that competitors could take advantage of low-hanging fruit. Yet this is simply not the case. Not only anecdotal evidence, but study after peer-reviewed study shows that, yes, blacks do make less money than whites. Not only anecdotal evidence, but study after study shows that blacks are discriminated against in housing.
So when you have blacks being discriminated against economically in other venues despite what the 'theory' says, and when you have gobs of anecdotal evidence that yes, they are being discriminated against in terms of credit and loans, well, then those relying on 'theory' have got some 'splainin to do.
I would think that they would realize by now that falling back on theory is a _liability_. Most definitely not an asset, and certainly not something to be proud of, let alone think that it trumps anecdotal evidence.
My question would be, is there any good reason to think that evidence of credit discrimination would be harder to suss out than discrimination in places like the labor market? It's not something I know anything about. I can understand how an economist would lean on an economic examination to provide the relevant evidence. But if you really wanted to prove/disprove the existence of the phenomenon in an empirical manner, how would someone go about doing it?
It's a case where one side has anecdotal evidence and the other has studies and basic economic theory.
Where are these relevant studies? None of the links provided are relevant to this specific claim. And individual companies do lots of things that are contrary to basic economic theory. I'm sorry, but saying "no companies do this, because doing it would violate basic economic theory" just isn't compelling.
"Which takes me back to the specific - Megan claims that people and institutions that do not behave in a way that will eventually erase these disparaties in service are behaving 'irrationally'. Are they? Or is there a good, selfish reason for this behaviour that hasn't elucidated quite yet? In any event, all she has is a theory with no data. And since there are many, many, many instances where 'theory' proclaimed that discrimination would go away and that it was 'irrational', yet the theory was completely, gloriously, astoundingly wrong, the default isn't that others have to prove the theory wrong (this wouldn't be true in any case, as a matter of good science), but that she has to prove her theory right."
1) You make a very good point with regard to economic assumptions regarding individuals but there's a huge difference between individuals and institutions (for better or worse). Ascribing a racially discriminatory motive to an institution is a big leap; in modern America, it would be very very difficult to keep this a secret from everyone considering who would need to be in on the plot (if you're a lawyer, find one programmer who worked on the racially discriminatory algorithm and you're a billionaire). If you don't keep it a secret from pretty much everyone, you're inviting a huge legal judgment (and having the press write article after article excoriating your company). There can be instances of corporations catering to the prejudices of their customers but that's a pretty different case.
2) You still rely on an unproven assumption that any differences in treatment between races is due solely to animus or preference. This ignores the reality that on almost any measurable characteristic (including credit worthiness), races do indeed differ.
3) In this specific instance (lending) if the racial bias were irrational all it would take is one lending firm to demonstrate this by lending to "traditionally under-served communities". Others would follow the profits. It's notable that no firm has done this yet.
Can I point out that, while not logically compelling, most of us live our lives according to anecdotal evidence? [...] Personally, if many black people say that they have experienced unusual difficulty in getting credit, then I'm inclined to believe them.
Wow. It's tough to imagine anything that this sort of thought process wouldn't support. The priests, commisars, and con men must love you...
Well, Dave, given that the argument on the other side is not exactly inundating us with evidence, only a discredited theory, and given that the discredited theory has made the same pronouncement in other aspects of economics, and been proven wrong by good solid studies as well as anecdotal evidence, my money's on the anecdoters.[1]
Do you have any reasons in hindsight as to why the theory was wrong in previous applications?
[1]That's not to say they're invariably right of course, only that thus far, theory alone has been a pretty poor predictor of the true situation. It might well be right this time. But I'm definitely going to need something besides econ 101 for verification.
I don't think anyone is suggesting that major lenders today have a formal or informal policy of giving African-American borrowers worse deals. Thus, there is no way for them to adjust a policy that they don't have.
Discrimination gets fed into the system mainly during the face-to-face meeting between agent and potential borrower; the biases of the bank's employees are the source of the problem. (And it may just be some of the employees, some of the time; we're talking about a statistical effect, yet one that is nevertheless perceived as a general conspiracy.)
I don't think it's irrational or a sign of diminished intelligence for someone to buck the system. People who are discriminated against are faced with a choice between self-interest and self-respect; there's no such thing as an irrational choice in that situation, unless it is irrational to have self-respect. The situation is, of course, tragic.
If you're one of the lucky people in society, you don't have to make these choices; and if you're really lucky, you get to claim selfish behavior as a social good and thus a source of pride.
Steve Johnson brings up some presentable arguments, but I still have more skepticism. Yet I can't claim to know exactly what he's getting at:
This may be true. But I'm not sure whether you're arguing for race-based determinism or whatnot. The IQ figures mentioned elsewhere are interesting, but so, too, is the observation that IQs increase in any group on generation-to-generation comparisons.
It may seem like a trivial point, but when we discuss average heights or weights or eye colors for different groups, we are dealing with yardsticks that don't substantially change over time. With IQ measurements, there are constant refinements and recalibrations. There's no ultimate way to account for and balance for testing bias. And cultural influences cannot be dismissed when we're talking about a minority that has developed its own cultural memes that include suspicion of authority.
I'm not trying to dismiss bad behavior or any type of lawbreaking, but the IQ lens -- a useless collectivist means of looking at groups rather than individuals -- provides a distorted picture, even though it may have a factual, accountable basis.
To illustrate this more anecdotally: You wouldn't tell some shut-in white senior citizen that statistics tell them that white strangers knocking at their door are less likely to be con artists or violent criminals than black strangers doing so. At an individual level, the collective data has virtually no useful value, I would argue.
"But if that were happening on a widespread basis, loans to minorities would be insanely profitable."
Auto insurers discriminate against men yet policies sold to men are not insanely profitable. So there is a hole in your argument. Much discrimnation is rational.
I would merely add that credit checks are not racist to the extent the results are based on actual credit history.
Blacks should be thrilled now that loan determination is based on algorythms and not the insight or whims of loan officers.
On the other hand, if assets are included in the method to determine credit worthiness, then one bumps up against historical issues.
As recently as the 1950's and 60's blacks were quite limited in all those things that would enable the transition of wealth from generatin to generation. As someone pointed out, each black generation starts with less assets on hand than a white person might.
I am not sure how you account for that though in determining credit. Clearly on a purely rational level, more assets means a better credit risk. But if a group of people were not able in the past to freely build assets, then what?
Probably no way around that, I guess.
As recently as the 1950's and 60's blacks were quite limited in all those things that would enable the transition of wealth from generatin to generation
Like marriage?
"As recently as the 1950's and 60's blacks were quite limited in all those things that would enable the transition of wealth from generatin to generation. As someone pointed out, each black generation starts with less assets on hand than a white person might."
I'm not sure that would be the case. At least currently, any person, white or black, can expect to be in their forties before they receive an inheritance. Except in relatively rare circumstances, most people, white and black, are starting from the same place in their twenties and thirties, I would think, from an appreciable asset standpoint.
MarkG - But when we discuss average heights or weights or eye colors for different groups, we are dealing with yardsticks that don't substantially change over time. With IQ measurements, there are constant refinements and recalibrations.
Are you sure about height? It was my understanding that height had some correlation with IQ. But that once malnutrition and ill health were factored out, the gap disappeared.
That could well be, and I don't think I'd be all that surprised. But I'll confess at the outset that this sense is based on thoroughly banal, trivial observations -- not at all academic or mathematically founded.
In sociolinguistics it has been observed that small groups of people develop their own dynamics in "construing" certain individuals as authorities or experts while relegating others to the margins. (The caveat here applies, as Megan's post on "winning the cocktail party" post suggested, individuals actively assert their own status in the group, too.) You can actually observe some signs of this dynamic on videos of people interacting and, in analyzing the video, trying to figure out the verbal and non-verbal signals people send one another in construing authoritative status.
Even more trivially, I would observe that there's a certain propensity for (vaguely defined) people to defer or otherwise lend greater weight to taller individuals. I can't remember where, but if I'm not mistaken, even economic studies have found income differentiation in the aggregate based on people's "tallness." Generally, the taller a person is, the more likely the person is to earn more and be promoted as a measure of workplace success.
We're social animals, and as such, we collectively produce order in small groups in very subtle ways which we're not always cognizant of. There's also (I say, going deep into the outfield of banal observations) the individual dynamic at work: We grow up looking up to parents and teachers, quite literally, such that people we naturally look up to are psychologically construed as authority figures that many of us might defer to by force of learned childhood behavioral habit.
There have also been trivial sociolinguistic studies showing how people change the way they speak when dealing with children -- the linguistic register of "caretaker talk" (I think), also familiar as "foreigner talk" when interlocutors assume that their counterparts lack a full command of the language of communication as non-native speakers. I would assume that the same choice of linguistic register applies when we speak to people we assume are "smart" and "authoritative." We don't "talk down to" the people "we look up to."
I also believe that individuals who are deferred to as experts authorities would generally tend to live up to the socially assigned expectations in whatever way they can. Here, I'm thinking along the lines of the teacher in the sixties who broke her elementary school class of white children into "brown-eyed people" and "blue-eyed people" to teach them about the impact of segregation and its attendant baggage of different expectations for the favored and disfavored groups. (I had to look that one up because her innovative experiment deserves mention, in my view: The teacher was Jane Elliott.)
But taller people are more likely to suffer from broken bones, if it's any consolation to the rest of us pipsqueaks. =D
I'm skipping to the end, here, so please forgive me if I'm repeating materials already covered.
As usual, Kathy G has some comments, and again as usual, she cites good research:
http://thegspot.typepad.com/blog/2008/07/by-kathy-g-a-ce.html#more