The anti-meritocracy of affirmative action

It’s well reported that black and Hispanic Americans have better chance to get into college with the same scores as whites, while Asians have less. Even the SAT difference is assigned. I was more curious how much difference this means to people with equal results.

The association of American Medical Colleges published its acceptance results for all four racial groups, in a form of table listing the GPA (how did the applicant perform in high school) and MCAT (how did the applicant perform on the college test). For example we see that 102 white people applied to med school with 3.0-3.19 GPA (OK-ish) and 5-14 MCAT (horrible) and not a single one was accepted.

Assuming meritocracy, race should have no effect on admissions, if two groups have the same GPA/MCAT, the admission results should be the same. I compared the grids of the different races, discarding fields where less than 50 applicants were present. Well, the table is very unequal:


If you have good scores, you get in, no matter your race. If your tests and high-school performances are mediocre, your chances to get in are slim – if you are white or Asian. But if you are black or Hispanic, your chances are much-much better. It’s strange that Hispanic advantage peaks at low scores. It’s probably because of historically black colleges. I mean that a dumb black will apply to such college but gets rejected because OK blacks get the places, while there are no historically Hispanic colleges, so a dumb Hispanic will apply to a majority-white college and … get admitted due to affirmative action. Someone should tell the horrible scoring blacks to go for majority-white colleges instead where the black competition is lower.

No need to say how damaging this will be to the future performance of professionals. I’m pretty sure that the skyrocketing stock markets and growing GDP aren’t because of actual Trump actions (they weren’t many), but because of the expectations that this affirmative action madness goes away and companies can hire competent people again.


Author: Gevlon

My blog:

11 thoughts on “The anti-meritocracy of affirmative action”

  1. @Gevlon
    >the expectations that this affirmative action madness goes away and companies can hire competent people again

    Trump can strike a minor (but symbolically significant) blow against affirmative action whenever he wants. He doesn’t need permission from Congress.

    This change would *not* immediately dispel AA admission policies at universities, of course. Many of them maintain AA policies in order to remain eligible for grants (including state and private sources, which aren’t bound by executive orders). Some instutitions are ideologically committed to the principle, and so they would retain their AA policies even if money was not a factor.

    It’s also important to note that the average business owner is *not* required to mirror the AA practices that you analyzed. Under current guidelines, he’s allowed to have a 99% male workforce — if he can show that the total pool of qualified workers (and/or qualified applicants) is 99% male. He’s *encouraged* to employ more women (e.g. by accepting underqualified female candidates and giving them on-the-job training) but he’s not *required* to do so. Educational institutions tend to go farther, because it’s a way to break out of a dismal cycle (black family is poor → black children have fewer educational options → black adults get worse jobs → black family is poor).

    If you want to quickly stamp out AA then you’d need to get the Supreme Court to rule that the whole thing is unconstitutional. That would be tricky. Stare decisis obviously applies, the underlying facts have not changed, and the ruling is relatively recent.

    >Assuming meritocracy
    I know that you didn’t mean this seriously, but I can’t avoid a chuckle when I see this phrase. It’s similar to “assume that all of the cows are spherical and frictionless … “


  2. Based on the PDFs, there were 59,267 students accepted into medical school out of 135,840 applicants. The entire total pool of non-white, non-Asian applicants was 25,853, which is less than the total number of Asian applicants alone (30k+). Let’s pick an arbitrary cutoff, say, MCAT 30+ and 3.2+ GPA. Anyone above that gets auto-accepted, since we’re an ideal meritocracy. The result:

    -Asians: 18,291
    -White: 36,901
    -Black/Latino/Native American: 4,013

    That actually adds up to 59,205, so there are enough seats at college for them all. Nevermind that we have 31% Asian medical students when they make up 5.6% of the actual population. This is pure meritocracy, so we won’t even consider how funneling opportunity for high-paying medical careers will impact people generationally, how they will buy more expensive homes that pay for more well-funded schools that might have a teensy-little impact on how the next generation fairs via GPA and MCAT. We’re plugging our ears and strictly looking at merit.

    …except, here’s the funny thing about merit and MCAT scores, from your own Wiki link:

    A recent study (2016), shows little to no correlation between MCAT scores and USMLE step 1 scores, as well as little to no correlation between MCAT scores and the NBME scores.

    In other words, one’s MCAT scores cannot predict one’s ability to pass the Board, two years into medical school. That’s the trouble with “meritocracy”: it depends on what you define as merit. There are no closed-book, multiple choice exams in the exam room, and we’re possibly wasting everyone’s time putting such an emphasis on MCAT and high school GPA when the actual job has multiple dimensions of merit. Part of that merit might actually derive from, say, a black patient feeling more at ease with a black doctor who, you know, still successfully graduated from medical school regardless of why/how he got in.

    P.S. The stock market rose based on expectations of tax cuts. Period.

    Liked by 1 person

  3. @Edwardqjones: expectations are just that: expectations.
    Businesses rightfully assumed that the likely Clinton presidency will push forward with AA, sooner or later demanding equal representation of identities regardless of qualifications. They also rightfully assumed that even if there won’t be formal policies, the constantly “progressing” society will demand it via boycotts or social media campaigns or their competitors get advantage by advertising that they are already more “diverse”. They had to factor these costs into their future profit calculations.

    With Trump winning, both expectations are gone and they can recalculate at worst with current situation, at best with dismissing existing regulations (so you can have 99% white males, no questions asked)

    The “Tricky” part will be solved in 2-3 years, since Grutter v Bollinger was a 5-4 decision and Ginsburg is 84, Breyer is 79.

    Now about the theoretical part. Even if we accept the lefty idea that the person is not responsible for his life and accept the “family is poor → children have fewer educational options → adults get worse jobs → family is poor” cycle, we have no reason to colorize it. Ergo, even if there is justification to AA children of all colors from poor families, there is absolutely no reason to give advantage to a black kid of a millionaire over a white kid of a trailer park redneck.


  4. @Azuriel: I have absolutely no problem with 31% Asians. I wouldn’t have problem with 100% Asians either. Actually the high % of Asians is a perfect proof that the “rich family, poor family” excuse is bullshit, as Asians are poorer on average than whites, so should perform worse instead of 6x better.

    The study is a selection bias fallacy. Ergo, it only proves that the MCAT of those who have high MCAT (and get admitted) is no longer the dominating factor in success. It doesn’t prove that those who had low MCAT (and got rejected) would be good doctors. Similarly it’s likely that those who write with 100% perfection (no typos and grammar mistakes) are not earning more than those who make 2-3 mistakes per page. But this fact does not mean that writing with 0% perfection (illiteracy) is OK.


  5. @Azuriel: sorry, missed the last line.
    If businesses would love Trump for tax cuts, they would have supported him. Yet the very same businesses that now publish cheerful predictions and buy everything that moves were paying 6 figures for a speech of Clinton.

    “We support Trump because we want tax cuts” is not even controversial, but nobody did that. This is because the true reason was unspeakable: “we (secretly) pray for Trump because he’ll save us from Affirmative Action”.


  6. @Gevlon
    >The “Tricky” part will be solved in 2-3 years, since Grutter v Bollinger was a 5-4 decision and Ginsburg is 84, Breyer is 79.

    That’s not how stare decisis works. You don’t get to relitigate arbitrary chunks of the country’s jurisprudence whenever a new SCOTUS justice takes the bench. You need to present new facts, or raise a novel argument, or show that the previous ruling is utterly incompatible with the values held by majorities of the people and of the states. You can’t just say “let’s have a do-over, except that my team gets +1 votes this time!”

    Read through for an instructive example. Conservatives had enough votes to shred Roe v Wade. That didn’t happen. Justices have political biases and party loyalty, but they also hold some respect for the institution that they serve. Each jurist would like to see his own sage wisdom reverberate through the ages; he doesn’t want to see all of his rulings tossed out if Hillary Clinton stuffs the court with a bunch of SJW crazies. Therefore he follows the Golden Rule. He *doesn’t* just try overturn every decision that he disagrees with. He respects precedent, and he trusts his successors to do the same.

    >even if there is justification to AA children of all colors from poor families

    Such progams exist, and have existed. Affirmative Action regulations didn’t just emerge spontaneously from the fevered brains of lefty morons in ivory towers. They were a response to real conditions, observed on the ground.

    Please do some research about the history of welfare programs in the United States. Most programs were ostensibly colorblind, and were focused entirely on need (i.e. family income, poverty threshold, food shortages, etc). But these programs were often *administered* in a racist way. The classic example is redlining under the FHA, which gave white families preferential access to affordable mortgages. For a more contemporary example, consider the erosion of minority voting access post-Shelby.

    And this isn’t purely about government action, either. Until 1964 it was legal for unions to exclude candidates on the basis of race, and many of them did so. Their motivation was partially economic (poor black guys will accept lower wages, thus driving down our standard of living!) and partially racist (black people are icky!). So even if you broke out of the cycle of poverty, pulled hard on your bootstraps and obtained a good education – you might still be blocked from getting a good job.

    Aiming for meritocracy is a noble goal. But it’s not reasonable to assume that racism (and its effects) exist entirely in the past. You can’t just pretend that we’ll *achieve* an untrammeled meritocracy if we remove all of these “obsolete” anti-racist policies. You need to do serious work, and *show* that the pernicious effects of AA outweigh its positives.

    Or you could ignore the consequentialist argument and rely entirely on deontology. In other words, you could try to revert American jurisprudence back to the Lochner era. But you should probably do some reading about that era first; the legal underpinnings were specious and the real-world consequences were often unpleasant.

    Liked by 1 person

  7. @edwardqjones: the new fact presented to the court is exactly the new fact that propelled Trump to the presidency: SJW-ing created exactly what it claimed to fight: an unequal society where “vulnerable minorities” are entitled to everything at the expense of the “majority”. Or simply speaking: the risk that the country slips back to the Jim Crow era is smaller than it becoming its inverse where whites, males, heterosexuals are formally oppressed. There are open voices demanding no votes for such groups. BLM locking down universities, denying the right to study to everyone based on banana peels is reality. Trying to discuss AA without mentioning BLM as a fact is like trying to discuss it in 1970 while acting like KKK isn’t there.

    I agree that colorblind meritocracy is unfair to those who were born with disadvantage. While I always believe that one can advance himself from his former self or his peers, I accept that one may not be able to reach someone who is worse than him but more privileged. But I also believe that any alternative of meritocracy is worse than it, as evidenced right here, droves of idiots admitted to med schools because of their color.


  8. @Gevlon
    If businesses would love Trump for tax cuts, they would have supported him. Yet the very same businesses that now publish cheerful predictions and buy everything that moves were paying 6 figures for a speech of Clinton.

    It’s shareholders (and investors) that love lower taxes, and who control share price, not necessarily the businesses themselves. Lower taxes = higher profit margins = more dividends = higher share prices.

    But I also believe that any alternative of meritocracy is worse than it, as evidenced right here, droves of idiots admitted to med schools because of their color.

    You haven’t actually demonstrated that. You have shown that AA allows more non-Asian minorities into medical school who have performed worse on a standardized test – which has no correlation to their ability pass the Board exam – than what might otherwise occur. Actual medical school and the Board will filter out any “idiots.” And you can’t even really argue that AA admissions are preventing “better doctors” from getting into med school, because the only indication we have that a given person is good is a test which doesn’t correlate with that conclusion.

    Again, this is the problem with meritocracy: how are we defining merit? If you want to present research that AA applicants end up having worse patient outcomes once they graduate med school, then present that. The only merit that should count is the one directly related to the job; standardized test scores are not (in this case) indicative of anything. Meanwhile, we have some (old) studies showing patients are more satisfied when seeing doctors of their own race. If you have conflicting studies, let’s see them.


  9. The really interesting consequence of AA is the following work discrimination of blacks and Hispanics.
    Worse students will make worse professionals, on average. Black graduates are on average worse than white graduates.
    If Universities didn’t use AA then all the graduates would be pretty much the same. Employers would have no problem with hiring blacks, as the field has been equalized by both entrance exams and regular Uni testing. As it is, employers can reasonably expect a black person to be worse at their job than an Asian.


  10. @Azuriel: if we assume that lower score, AA blacks won’t be worse doctors, then we should assume that lower score, rejected whites would also be. That – at best – shows that GPA and MCAT are bad test (this is something I gladly agree with, GPA is the average of ALL grades, I’m sure that a student who is reading biology books all day but barely pass History and Literature will be a better doctor than the guy who learns everything the teachers feed him). But currently GPA and MCAT is all we have and there is no reason to prioritize blacks over Asians (funnily, whites, especially non-Jew whites would be winners of a strict “every group must have equal share” system)

    Patients being satisfied by doctors is irrelevant, or very emphatic snake oil magicians will be the doctors.

    However I did not answer why AA is so bad. All I demonstrated that it’s a wasteful and unfair program (1000 class AA university is equal to a 700 students fair university + 300 students university that admits only low score minorities and then they drop out before graduation on government money). This is simply welfare. The real problem will come Monday (and I wouldn’t even think of it without your comments, thank you, “unfair is wasteful” was enough for me)

    @Stawek: not necessarily if the universities fail the bad students and if employers ask for work experience (ergo forcing graduates to do another year of unpaid internship)


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s