« Affirmative Action--Posner's Comment | Main | Response on Affirmative Action-BECKER »



Feed You can follow this conversation by subscribing to the comment feed for this post.

Jake Sinclair

Becker's argument is fine. But then one can go on to ask why affirmative action, rather than other policies, might have been adopted as a strategy to address racial disparities in employment and education. Part of the answer could be that affirmative action imposes large costs on only a few -- the better qualified applicants who are denied the positions they merited. Everyone else pays a cost when less deserving individuals fill slots that are in demand, but these costs are hidden and probably small. Affirmative action, then, is a relatively cheap way for society to pretend to address racial disparities. Becker's preferred methods, like better early childhood education, are more expensive up-front than affirmative action, which requires little tax money because the costs are chiefly borne by private individuals. Viewed from the standpoint of distributive justice, affirmative action seems terribly unfair and also rather cheap.

Leftists can endorse this view: see Michael Walzer's discussion on "the reservation of office" (quotas) in Spheres of Justice.

Bill Korner

On my view, there are two rationales for affirmative action. They are analytically quite distinct but in practice sometimes overlapping.

The first rationale is providing opportunities for individuals of underrepresented backgrounds. It focuses on the benefit to those individuals, as well as general ìmoraleî (as Posner says) among members of that group. Curiously this is the only one discussed in Becker and Posnerís posts.

The second rationale, is that diversity can improve the quality of organizations or institutions. For example, diversity at universities exposes students in dominant groups and ones in certain ìdisprivilegedî perspectives to each other. Or diversity on corporate boards of directors or among management might conceivably improve profitability or have some other advantage. (The scholar Lynn Dallas thinks so at any rate.) Here the focus is on benefits to some ìcommunityî that do not involve redress.

The first and second rationales overlap in academic settings and certain employment contexts. The latter does not seem to apply much in the context of preferences in awarding contracts.

In our affirmative action jurisprudence, the two rationales are not very well distinguished analytically. But, to the extent that they are, I believe that the latter has garnered more approval in our jurisprudence.

Given that I am not a federal judge, I guess that mine is an ìidiosyncraticî take on the jurisprudence. But, though I take my point of departure from that jurisprudence, I am not intending to make the kind of argument that can be dismissed by showing Iíve forgot my constitutional law.

Iím just saying that, in my opinion, the second rationale is by far the stronger, specially since it thinks in terms of benefits to each and everyone (thus promoting unity) rather than to individual benefit and group grievance.

Furthermore, I think the authorsí arguments about meritocracy entirely fail to address the second point. Their assumption is, on the contrary, that meritocracy is so powerfully important a concept that loyalty to it should prevent us from considering any other effects that diversity in institutions and organizations may have on our country. That seems to be a very, very big assumption and quite questionable.

[This is not to say that right wing legal economists would not differ markedly from left wing ones (if there are any) on the questions of (1) what meritocracy means and (2) if we have one. I think that they do or would, because left wing legal economics would be distinguished by a sensitivity to what effects wealth distribution may have on evaluating merit and meritocracy.]

Richard Mason

[S]ome states... admit applicants to state universities and colleges if they rank in the top 10 per cent of their high school class. This is disguised affirmative action because schools with favored minority groups typically have much worse students than other schools, so it is considerably easier to rank in the top 10 per cent of the lower quality mainly minority schools.

This practice is not merely a disguise since it does improve on straight race-based affirmative action in a couple of ways. First of all, anyone is free in principle to benefit from this kind of affirmative action by moving to a poorer school district. But I suspect almost no parents would consider that to really be in the educational interest of their children, which leads to the second point, that "school attended" is probably a better indicator than "race" of which students are disadvantaged and (arguably) in need of special consideration. And third, this practice only has an effect so long as there is an existing difference in quality between schools; it would have no effect merely because some school had been harmed or deprived in the past.

I also observe that, for better or worse, geography-based affirmative action is a feature of our system of government. (Rhode Island gets two Senators regardless of whether they are more or less well qualified than the other Senators.) If a neighborhood is entitled to send a certain number of representatives to the state assembly, it doesn't seem so strange that it should be entitled to send a certain number of kids to the state university.


"Iím just saying that, in my opinion, the second rationale is by far the stronger, specially since it thinks in terms of benefits to each and everyone (thus promoting unity) rather than to individual benefit and group grievance."

Except nobody really believes the diversity rationale. Why aren't they looking at promoting religious or political diversity if they really buy into this "diversity" sophistry? That and the idea is kind of racist. Black people apparently have a certain perspective and white people have another, and Jewish people yet another. Oh, and we all know Hispanics have something nobody else can bring to the table. That or one thinks diversity of skin color is beneficial per se. I am sure the sight of brown and black skin of various hues warms the heart of many professors. That is reason enough to keep it--aesthetics.

"That seems to be a very, very big assumption and quite questionable."

What seems to be a big assumption and is quite questionable is that diversity has any educational benefits whatsoever. In addition, it is quite racist. Maybe before buying into this "compelling" state interest, you know, you might offer some evidence.

Jews comprise approximately 2% of the population but make up 25% of the nation's law professors. In the name of diversity, are you willing to engage in discrimination against Jewish law professors in employment? I would hope not. But if you take this "diversity" business seriously, that is precisely where it leads...


Richard Mason makes a good point. Anything which targets the disadvantaged more efficiently than the current system is a major plus. I don't get the justice of sending a black kid with professional parents to Harvard before a poor white kid from rural America. What horrible injustice is affirmative action correcting by that? Oh, yeah, diversity of skin color comes before diversity of subtance.


Please don't bring words like "diversity" into this argument. The multiculturalism and diversity movement argument holds no water. It is the last resort of the moral relativist. The belief is that people coming from different backgrounds with different worldviews and cultures can more aptly solve problems. I think that view is completely bereft of intellectual rigor. It is always the best ideas that win out in the marketplace of ideas. Why do we suddenly need diversity to ensure good ideas, when we did just fine without for, oh, 50,000 years?


"Except nobody really believes the diversity rationale."

Yes, I really believe the diversity rationale.
many of my friends and associates around the world do as well. I am more interested in diversity of viewpoint but I am also willing to use race, gender, sexual orientation, religion, class, or geographic origin as a proxy for "will have a unique outlook" because it seems to work in practice.

"It is always the best ideas that win out in the marketplace of ideas."

Hate to point this out, but the marketplace of ideas is diverse. (Though individual forums may be rigorously self-polarizing)

So far every commentator on this side of the blog including Becker has missed an important possible grounding for Affirmative Action. If we agree with Posner that African Americans face a disadvantaged position on average vs. other races, then there is also a problem with standardized tests or any other "meritocratic" measuring stick that is applied to both groups equally.

The LSAT does not correct for the experience of growing up with fewer books in the house and more crime on the streets. The children who live in these impoverished urban areas can not be blamed for their environment. Yet they are judged on the test in exactly the same manner as people who didn't face those extra challenges.

I doubt that anyone here would disagree that growing up in the Bronx or East LA has a negative effect on one's education. Therefore it has a negative effect on one's test scores. The test is NOT "standardized" because it does not take that into account.

So Aff. Action programs DO take the no-fault environmental disadvantages into account.

"so it is considerably easier to rank in the top 10 per cent of the lower quality mainly minority schools."

I couldn't disagree with that statement more. The top ten percent in those schools represent students that have worked hard, overcome the environment, and bettered their fellows. If you had started them in the "higher quality" school they might still be in the top 10%.

If you take a more holistic view of "performance", then a top student is a top student. The top student in Watts might excell at thinking under pressure and distraction, the top student at Phillips Exeter might excell at differentiating brands of Khaki and formally deconstructing arguments. I'd rather have the Watts kid working with me.

So in a way, Affirmative Action is a necessary result of the narrow "standardized" way that Americans define their meritocracy. In order to generate standardized evaluation criteria, we narrow the evaluation and end up preferencing certain kinds of success.


"What seems to be a big assumption and is quite questionable is that diversity has any educational benefits whatsoever. In addition, it is quite racist. Maybe before buying into this "compelling" state interest, you know, you might offer some evidence"

I learned a great deal about poverty, homelessness, and their effects from talking to people living those things. As a result I am often struck by how little commentators seem to know about the issues when they pontificate from their air-conditioned homes and studios.

All I know about how to be a good leader/manager in an office environment was learned from an Iranian boss who frequently incorporated references to Eastern culture.

My girlfriend reports that classrooms with mainstreamed "special needs" children are more sympathetic, supportive, and helpful to each other than classrooms without. It is a common goal of elementary educators to instill these character traits.

The most interesting days in law school were the ones in which 10 or 20 students spoke up with their alternate interpretations.

You will doubtless attack my evidence as anecdotal. Perhaps when you do you will include some evidence in your preferred format.


Becker is right on to say that aggressively prohibiting discrimination and spending to help children at an early age is the ideal approach.
That isn't an argument against Aff. Action though if it can be shown that Aff. Action is also beneficial.


But Corey, that is exactly the point. Affirmative action is not beneficial. Read Thomas Sowell's compendium on it. He engaged in original research on affirmative action, not just here but in its manifestations in Malaysia and China and other countries, in addition to black students at Cornell. His research shows affirmative action places minority students in environments that are not suited to their academic skills, pretty much exactly what Dr. Becker said, and that if we got rid of affirmative action, you would see black graduation rates rise. His theories have been shown to be correct in California public universities.

Also, what do you mean by "beneficial"? More minorities graduating with degrees, or simply more minorities in college? I prefer the first, and Dr. Sowell's arguments skewer the current system when it comes to providing that.



Maybe I could take you seriously if there were homeless, poverty, or alternate philosophy affirmative action. No, there is race based affirmative action. Universities are doing NOTHING to promote any other sort of diversity (except maybe gender, and that is largely unncesssary except in engineering/physics/math programs). If diversity is the real reason, why not discriminate to achieve religious, cultural, and political diversity? I suspect there are no efforts underway because the diversity proponents don't really believe in it.

I have a question for every one who believes in the diversity rationale: Are you willing to discriminate against Jewish professors to make room for other, under-represented groups? And if that is not OK, then why is OK to discriminate against whites and Asians generally? If one accepts the diversity rationale, there is really no limit to what can be done to promote the "diverse experience."

All other things equal, sure diversity of experience and backgrounds can't be a bad thing. But does it justify discriminating, on the basis of race no less, against more qualifed applicants? The civil rights movement was formulated around the concept that race didn't matter, that individuals should be "judged by the content of their character, not by the color of their skin." Yet affirmative action rejects this noble notion, instead arguing that race does matter, and that it is a legitimate basis for discrimination (if you're white or Asian).


Testing gap:

Under affirmative action, we would expect this gap to widen compared to a system of racial neutrality. Blacks, Hispanics and other recipients of affirmative action know that they do not have to acheive the same scores as whites and Asians, so there is less incentive to score well. Why spend hundreds of hours in preparation, or in attending to one's studies assiduously, when one knows they are treated under a more lax standard where they can more easily gain admission ot their school of choice? One would expect the recipients of affirmative action of all abilities to under acheive in standardized testing because of this lowered standard.


I apologise for the multiple posts...

For those who like evidence and reason:

Becker talked about research [from UCLA] that sugggests "preferenced" students cluster
at the bottom of the law school class:

Other research by Prof. Henderson at Indiana has shown how LSAT performance is strongly correlated to performance on timed law school exams, but is weakly or not correlated to performance on other kinds of exams (take home, non-timed, oral). In essense, a measure of merit used for admission to school has evolved to be duplicative of the measure of merit most often used for performance in school.

That explains why students preferenced into a school might duplicate the same performance gap while in the school. But legal practice does not resemble a timed law school essay exam. There is no reason to suspect the performance gap to extend beyond law school into legal practice absent discrimination there. Especially given the weak or nonexistant correlation between LSAT score and exams types that DO resemble legal practice.


"His research shows affirmative action places minority students in environments that are not suited to their academic skills"

But, as the research I mentioned above (as you were writing your post) indicates, the academic environments are duplicating the placement criteria, and neither are necessarily correlated to the skills necessary for success afterwards.
Perhaps academic environments could be made more "fair" by applying some diversity to the type of performance evaluations they do.

"Are you willing to discriminate against Jewish professors to make room for other, under-represented groups?"

Anyone who supports Aff. Action and thinks about it would note that it could "discriminate" against groups represented above their "natural" statistical population. That would include whites and Jewish professors.

If you are committed to the belief and evidence that race does not pre-determine merit in a non-discriminatory world, then you would expect the law of averages to work so that people were represented in proportion to the size of their group.

"If diversity is the real reason, why not discriminate to achieve religious, cultural, and political diversity?"

Yes, why not. I am for it. I agree that there are few efforts underway.

"All other things equal, sure diversity of experience and backgrounds can't be a bad thing. But does it justify discriminating, on the basis of race no less, against more qualifed applicants?"

You've begged the question by saying "more qualified". There are inequalities in background (re wealth) that don't relate to merit yet do affect the test. The test doesn't take these into account so its determination that I am more qualified than the student who struggled to the top of a blighted school is suspect.

And again, if the test duplicates the grading system it admits for without correlating to alternate ways of grading, then the test does no work at all.

But yes, if diveristy is a positive good, then in the language of cost benefit so popular on this blog, it does justify some lesser degree of "harm".


"If you are committed to the belief and evidence that race does not pre-determine merit in a non-discriminatory world, then you would expect the law of averages to work so that people were represented in proportion to the size of their group."

Jews and Asians have succeeded in spite of their race, not because of it. They are over-represented in the professions and earn incomes that exceed the average for whites. Why is that? Culture plays a role here, not race (though race correlates with culture). I suggest you read some of Thomas Sowell's work on the topic.


Diversity is a weak rationale for affirmative action (that for historical reasons is now at the forefront). Yes, all else being relatively equal, let's take the guy from Montana and not Tennessee since we have three from Tennessee already. But diversity isn't meant to be the tail that wags the dog.

Affirmative action, when it produces the whopping favoritism that it does for African-Americans (I don't fear so much for most other groups, especially women, since they don't need much special help) will effectively keep them down longer than doing nothing. If you tell a group they don't have to perform as well as other to be rewarded, there is no reason for them to apply themselves as much. While the problems of blacks in America goes deeper than that, and certainly there is the issue of discrimination, it is nevertheless true that no two-tiered system where less is expected of group A than group B will lead group A to perform better in the long run.

Matthew Collin

This is just an idea that came to mind while reading the other comments, and it depends entirely on the assumption that SAT performance is a weak predictor of university success (note that universities admit based on a collective score, and do not take Verbal and Quantitative sections into account separately, thus a person who underperforms at math but gets a 1300 can get into the same engineering program that a person who underperforms at English and also gets a 1300.)

Assuming that the correlation between performance on the SAT and actual undergraduate performance is less than perfect, then there must be a cohort of African American students who desire admission to top schools, yet know they cannot 'make the grade.' As the expected chance of admission is 0, their performance will match the admission standards of the next best school.

However, if top schools lower the bar just a little, then these students will react to the rising statistical probability of admission by working harder and scoring better. Due to the inaccuracy of the SAT, these students could work just as hard as the rest of the student population. However, this argument could be applied to all underperforming students, not just African Americans.

Matthew Collin

Jon Stewart of 'The Daily Show' once made a joke that the white student who was denied admission to Princeton to make room for an African-American admission had to rely on his fallback: Stanford. Without realizing it, he brings up another interesting point:

Applying to the top schools in this country is not free. If you have two students with the same ability but separate incomes, the richer of the two will be more likely to be admitted into a top school because he is able to pay for more applications than the poorer of the two. Again, the largest problem with this argument is that it promotes affirmative action aimed at the poor, not at ethnic minorities.

Bill Korner

Diversity is obviously a difficult concept, one that has to be about more than head counting with respect to race, creed, ideology, ancestry, etc. I have no grand theory on how to bring it about in the law school context or any other.

But the fact that "diversity" as a criterion is not easy to apply or is hotly contested does not make it unimportant. Nor can the concept be said to be discredited, although the difficulties in agreeing on how to use it put those who insist on using it on the defensive. This can rather be evidence that it is an important concept the fleshing out of which is a crucial if difficult conversation. In this case, the heat of the debate about diversity is shows that the important conversation is not being carried on successfully.

Neither, for that matter, does the fact that a criterion is relatively easy to apply (e.g. LSAT scores) make it appropriate. At most the former are inconvenient and the latter are convenient. I agree with those who say the LSAT should be more controversial than it is. But one must sympathize as well with the plight of admissions committees.

I'm ambivalent about affirmative action because it means so many different things, from quotas to the diligence in searching that Becker supports but does not seem to consider affirmative action.

One thing I am confident of is that to date it has been something of a BAND-AID. Some of Posner's discussion also suggests this BAND-AID view: That affirmative action usefully provides a sense of hope but does not really remedy past discrimination and the attendant reason(s) for lower levels of social advancement among blacks. I would agree and also think that a robust school voucher plan of the kind advocated at one time by Robert Reich would be way less of a BAND-AID (the fact that that term of derision was invented for vouchers notwithstanding).

Patrick R. Sullivan

'...no two-tiered system where less is expected of group A than group B will lead group A to perform better in the long run.'

Exactly. Which why AA is so pernicious, it teaches the wrong things to people who desperately need to know what Shakespeare knew 400 years ago; sweet are the uses of adversity.

Read Tom Sowell's autobiographical 'Black Education, Myths and Tragedies', to see why he says he thanks God that he grew up before AA.


While agreeing with the arguments against affirmative action as it is practiced in the U.S., i'd like to quickly point out that there are no affirmative actions in place in western europe, contrary to what Posner wrote. Only law enforcement tends to 'try to hire members of minorities as polive officers', but no specific numbers are attached to this policy. Continental European (not U.K.) policy focusses on lowering financial hurdles during education. For example, in Belgium the government strives to make education as cheap as possible (i.e. all wages are paid by the state ... lower grade and high schools are free to choose their religious orientation). Second example: the inscription fees in Belgium cost max. $500 per year at university (poor people can enter for free -- again there are christian as well as state universities). No entrance exams are required for any courses ... although some studies such as medicine where the supply has overshot demand temporarily organize entrance exams. The reasoning is that cheap, accessible education from the earliest age should make affirmative action at university entrance and during the subsequent career avoidable. Further underlying reasonings are that affirmative action at those later stages are too little, too late, and that lowering the hurdles to high quality education according to one's income is a more racially neutral policy.
The recent NYTimes articles on class in America suggested that social mobility in the United States is lower than in many western-european countries (incl. the UK, which us continental europeans still see as very much class-oriented), so we might be on to something.
The downside? Why, high taxes of course! 40% of the Belgian national budget goes to these educational subsidies. Although I have to say that our highest tax bracket is 55% (for incomes above 62,790 euro) -- http://www.taxup.be/nl/rates/tarieven/tar3.htm --
so our tax levels aren't as 'dramatic' as some Americans think. Especially since these taxes also include a National Health Service.

Ken Zimmerman

Meritocracy!! There is an extensive research literature on what it is and is not, and the great harm and little good it has done. READ IT!!

You complain about "affirmative action" as "us[ing] lower standards for African Americans and members of various other minority groups than for white males in determining whether they are promoted to higher level jobs in private business or government, admitted to better universities, and in other situations." I agree the standards are different. Whether they are "lower" is a very complicated social issue about who and how standards are defined. After all, the rich and powerful have always had "different" standards for all the things you mention. Do you really believe G.W. Bush was admitted and graduated from college because he "merited" it? I can tell you from personal experience the answer is no. But it is the same for many other "identified" groups -- children of alumni, at religiously based schools those very committed to the faith, and great sports stars. So "affirmative action" in the US is not new. Its just new for the so-called minority groups you choose to mention. And that brings up the issue of why you only mentioned "minority group" affirmative action rather the other pre-existing types I mention. Could it be you are biased??

And this whole notion of opposing "interest group politics" you bring up is totally disconnected from reality. The US is and has been always an interest group country. The only things that appear to have changed is that interest groups previously excluded from political power are beginning to find ways to gather it and those who hold the majority of political and economic power in the US are finding new and clever ways to legitimize their positions that are really, in your words, "bad for a country like the United States."

Why don't you turn some of your intellect and wit to protecting democracy in the US. Affirmative action is a small pimple, if that, compared to the current attack on US democracy (even as weak as that democracy has been historically).


I see two major problems with the diversity argument for affirmative action:

1. If diversity of viewpoints is the only justification for affirmative action college admissions, there is absolutely no reason for the proportion of minority students admitted under affirmative action to have any correlation with the proportion of minority students in the general population. If Eskimos make up 0.1% of the student population and that's sufficient to represent the Eskimo point of view, then why would we need African Americans to make up more than 0.1% of the population to represent the African American point of view?

2. A number of comments have already pointing out that the diversity argument, if we really believe it, suggests that we should seek to recruit students with a broad range of political and religious views. That's true, but a better argument would be that seeking a diverse student body requires us to recruit a lot more international students than we currently do. Does anyone honestly believe that the background of an African American contributes more to a school's cultural diversity than a student from another country?

Bill Sippel

"No man differs more from another than he does from himself at a different time."

I cannot remember the philosopher who said this, but I did not expect to see the difference in the same paragraph--to be precise, to say that meritocracy should be the rule and then also support (or at least try to justify)legacy selection to a university.

This reminds me of some recent observations of my wife, a reference librarian at a public library: that a good number of city employees were related--mothers, fathers, nephews and nieces--all in various departments (police, fire, library, city hall) throughout the city. A coicindence--I don't think so.

We need to vigilent against all forms of discrimination and unmerited preference. Unfortunately, this is not on the agenda. Nor are legacy candidates or recipients the ones to lead the charge.

michael persoon

I believe that as far as "diversity" goes, the Sup. Ct. did a good job in the Michigan cases, where the law school policy allowed for individual treatment (and was legal) while the undergrad poicy did not allow for individual treatment (and was thus illegal). Where a person is treated as an individual rather than as a member of a class, policies preferring traits that certain individuals have can be used. Among them: gpa, SAT, artistic ability, athletic ability, whether your parents are alumni, ethnic traits. Some of these are "merit-based" categories of characteristics that individuals my exhibit (depending on your views on nature vs. nurture!) and some are "inherent, immutable" characteristics.

Additionally, while recognizing the existence of data that suggests AA harms minority groups more than it helps, AA can also be viewed as a correction to a market failure. My contention to this point is as follows:
1) For every position (job, school, etc.) there are more qualified applicants than positions.
2) Where qualifications are equal, other measures are used to allocate the scarce position, including things such as how well you fit in the office environment and what sports team you like (non-merit based).
3) That all things equal, a person will hire someone for a scarce position who has more things in common with him, i.e. the person with whom there is the strongest in-group identification with.
4) Due to our particular history, and government sponsored institutions including, but not limited to, slavery and Jim Crow (anti-market forces acting to subsidize a particular class), certain groups have been under-represented in the hiring market.
5) Since certain groups have been under-represented (as a result of anti-market forces) in the hiring market, those same groups will be under-represented in the hired class. See (3).
6) Therefore, limited application of affirmitive action to QUALIFIED individuals (as many, if not most AA policies require) directed to correct this market inefficiency is justified (as a corrective matter).

The comments to this entry are closed.

Become a Fan

May 2014

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31