Economists and biology

Naturally, economists know a lot about human biology and evolution, just as civil engineers have to know about the properties of timber, concrete and steel. They have a good grounding in psychometrics, behavioral genetics, and quantitative genetics – how else could they do their job? Populations vary in traits that play key roles in economic activity and growth – in intelligence, asabiya, savings propensity, etc – you have to be aware of that variation, else whole continents would be economic mysteries. In the same way they know that those observed differences are a product of selection – which means economic historians think seriously about psychometric changes over time and their consequences, such as the Industrial Revolution. That kind of analysis helps predict where modern economic institutions can be successfully introduced, and where they cannot.

Yet even Jove nods. Sometimes even tenured professors make serious errors on fairly elementary topics. Like anyone else who has made a mental typo, they welcome polite correction.

Deirdre McCloskey has a new book out: Bourgeois Equality: How Ideas, Not Capital or Institutions, Enriched the World. I’m sure that there are many good things in it. But McCloskey makes a significant error in talking about genetics. “Know also a remarkable likelihood in our future. Begin with the sober scientific fact that sub-Saharan Africa has great genetic diversity, at any rate by the standard of the narrow genetic endowment of the ancestors of the rest of us, the small part of of the race of Homo sapiens that left Mother Africa in dribs and drabs after about 70,000 BCE…. Any gene-influenced activity is therefore going to have more African extremes. The naturally tallest people and the naturally shortest people, for example, are in sub-Saharan Africa. The naturally quickest long-distance runners are in East Africa. The best basketball players descend from West Africans. In other words, below the Sahara the top end of the distribution of human abilities – physical and intellectual and artistic – is unusually thick. …

The upshot? Genetic diversity in a rich Africa will yield a crop of geniuses unprecedented in world history. In a century or so the leading scientists and artists in the world will be black – at any rate if the diversity is as large in gene expression and social relevance as it is in, say, height or running ability. ”

So by this argument that the most cold-tolerant Africans must be more cold-tolerant than Eskimos: but they’re not. The most altitude-tolerant Africans must be more altitude-tolerant than Tibetans – but they’re not. McCloskey is thinking that a turn to free markets will make Africa rich, and that will give educational opportunities to Africans now denied them – but a fair-sized mostly-African population already lives in the United States, a population that is already much more prosperous than sub-Saharan Africans. How are they doing? How many geniuses are they producing?

The whole argument is flawed. Overall genetic variation is mostly in neutral loci. By itself it tells you nothing about any particular trait. Europeans do have less overall genetic variation than sub-Saharan Africans (~20% less), but they show more variation in hair color and eye color than Africans.

Essentially every domesticated species has less genetic variation than its wild progenitor. Dogs have less genetic variation than wolves. So, does this mean that the tallest wolf is taller than any dog? No – the tallest Great Danes are taller than any wolf. The heaviest mastiffs are heavier than any wolf. Chihuahua are the smallest. Greyhounds are faster than wolves (by a little).

Thoroughbred horses have little genetic variation – their effective population size is under 100. Tarpans, the wild ancestor of domesticated horses, are extinct, but there certainly are horse breeds with much more genetic variation than Thoroughbreds. But they’re slower.

What matters is the frequency of alleles that influence a trait, not overall genetic diversity. If, for example, the variants that tend to boost educational achievement (some of which were found in the just-released Nature study) were on average less common in sub-Saharan Africans than in Europeans – says 5% less common – Africans would tend to do less well in school. Like they actually do. Now Africa is a big place, and some groups are genetically quite distinct from others. Bushmen are genetically more distant from the Bantu than the Bantu are from Chinese. Some African populations might have experienced selective pressures that were more (or less) favorable for intelligence than others. Is there evidence, either in test scores or cultural accomplishment (better than any test), that some African populations may have smarts comparable with, or better than, people in Switzerland or Holland or Scotland?

No. There is no such evidence.

How many real geniuses out of African populations – people like Fermat, or Riemann, or Gauss, or Laplace? Newton? Maxwell? Gibbs? None. How many of the top 1000 mathematicians in the 20th century were of sub-Saharan ancestry? One, perhaps?

McCloskey’s prediction of an efflorescence of African genius is based on an incorrect understanding of how quantitative traits work. It’s also in that interesting, ever-popular class of theories that are contradicted by everything that has ever happened in the history of the world.

This entry was posted in Economics, Genetics. Bookmark the permalink.

134 Responses to Economists and biology

  1. Rifleman says:

    McCloskey’s prediction of an efflorescence of African genius is based on an incorrect understanding of how quantitative traits work.

    Well, you know McCloskey is a dude who thinks he’s a woman so what do you expect.

    Add to the dog/wolf issue that Kangal dogs have a bite massively more powerful than any wolf. Also that despite their greater genetic diversity wolves are useless when it comes to learning anything other than being a wolf.

    They aren’t going to go into battle like Dutch Shepards or Malinois or sniff for dead bodies or drugs or snakes on Guam. Or fight to the death like an American pit bull or be permanently child-like like a Papillion.

    Also the strongest men in the world are found to come generally from northern Europe to eastern Europe.

    Little Iceland and some Slavic countries have probably produced more super strongmen than all of Africa.

    How much genetic diversity was found among Hungarian Jews, a population that produced plenty of genetic freaks for raw genius.

    • Jim says:

      When I lived on Guam as a kid I don’t remember that they had any poisonous snakes but maybe they’ve been introduced. They could have used dogs to sniff for explosives though. I remember once when I was in the sixth grade one of my classmates came ambling along with a grenade that he had found lying around outside. Our teacher quickly confisticated it. He wouldn’t allow us to have any fun.

  2. airgap says:

    Who’s the one, David Blackwell?

    • AnonymousCoward says:

      Perhaps “Black Science Genius Guy”.

      He’s famous, and has been in a lot of Hollywood productions.

    • Jim says:

      He’s certainly very distinguished and no doubt could be considered among the top 1000 American mathematicians of the twentieth century but squeezing him into a list of the world’s top 1000 mathemticians of the twentieth? I don’t think so.

      The problem with making up a list of the top 1000 mathematicians of the twentieth century is that pretty soon you have more than 1000 names.

    • sprfls says:

      That’s the most Jewish-looking black guy I’ve ever seen.

      • MawBTS says:

        Many high-performing blacks are actually “blews”.

        The Baltimore Ravens have a black football guard who’s something of a math prodigy. He’s authored papers and is now pursuing a PhD at MIT. His surname is Urschel and his bone structure looks almost comically Jewish. I’m guessing substantial Ashkenazic ancestry.

        • Urschel’s having a biography ghost-written. A friend in publishing alerted me to the fact that Urschel’s agent was hunting for a ghostwriter, someone with a math background. “Why don’t you apply, Derb?” my friend asked. I thought it would be fun, so I sent in a résumé.

          Didn’t hear back.

        • ziel says:

          He looks like Paul Simon (the songwriter, not the Senator). Of course he’s an offensive lineman.

  3. Peter Johnson says:

    Economists can get in trouble with their colleagues and universities if they publish any results linking genetics to economic outcomes, particularly if (as is often true) there is any potential race linkage. It is also very difficult to publish that type of research in mainstream journals. Oded Galor sort-of-manages to some extent, but he uses tricks to sanitize his results (such as describing results as “culturally inherited” when his findings obviously reflect mostly reflect genetic inheritance). It is difficult and the economics profession is cowardly and/or blinded by mainstream ideology.

    • No, no, no. Ashraf-Galor argues that genetic diversity correlates with economic development (though in an inverted U-shaped way), so their argument is subject to the same critique as Greg Cochran just made about McCloskey.

      At first Ashraf-Galor implied they were making a genetics discovery (!!!) by arguing that the correlation was causal (genotypical diversity caused all kinds of phenotypical diversity), and some of their students showed up at this very blog to continue to make that argument. But A & G appear to have changed tacks at some point and they (seem to now) largely argue cultural diversity sort of mirrors genetic diversity (which makes their reasoning more like Spolaore-Wacziarg). And now they have a new paper coming out in which they do a “falsification test” on whether genetics can explain some of the variation on economic development. They look for statistical associations between “distance to the genetic frontier” and various measures of economic development, and they conclude no. Hahaha !

  4. And for pretty much the same reason, the “falsification test” performed by Ashraf & Galor in section 5 of the latest survey paper “Macrogenoeconomics of comparative development” ( forthcoming in the Journal of Economic Literature) makes no sense at all.

    “Genetic distance to the frontier possesses no statistically discernible relationship with the level of economic development”. Hahaha !

  5. reiner Tor says:

    But that would be racist to suppose that Africans are less intelligent. Once you eliminate that requirement, you’re left with their supposedly higher variance. The only problem is that the theory is quite ridiculous.

  6. ckp says:

    The German government seems to be taking the psychometric literature to heart. I’m sure the results of this programme will be acted upon in a thoughtful, rational manner:

  7. Peter Johnson says:

    Oded Galor carefully wraps his research into politically-correct packages, but I wonder if even he (let alone his typical reader) believes the politically-correct spin he puts on them. At least he has managed to get some articles linking genetics (albeit genetic diversity not actual race-related differences) into major economics journals. That might eventually open the floodgates to honest debate in the profession on the evidence about race differences and economic outcomes (or maybe not). Economists are cowardly and blinded by ideology but they are not all stupid.

  8. Jim says:

    Who would be the individual of Sub-Saharan descent who would be among the tup 1000 mathematicians of the twentieth century?

  9. The Z Blog says:

    The market for fawning over sub-Saharan populations for a wide range of reasons is lucrative. The market for objective analysis of sub-Saharan populations is nearly non-existent. If you want to write books about people, you better understand the market place.

    At the same time, many economists get cause and effect backwards. They are convinced that economics drives culture. Arrange the economy in just the right way and you get the ideal culture with ideal citizens. It’s the old gag about a man being good with a hammer seeing the world as a nail.

    That said, the population explosion in sub-Saharan Africa is probably the most important topic facing civilization since the Huns began pushing Goths into the Roman Empire.

    • gcochran9 says:

      ” the most important topic facing civilization since the Huns began pushing Goths into the Roman Empire.”

      Probably not – they’re not militarily powerful, nor are they likely to become so. The threat of such immigration reminds me of a scene in a short story by Cyril Kornbluth – The Cosmic Expense Account. A little old lady has accidentally acquired vast psychic powers, and her influence is expanding at the rate of a mile per day. Within the Plague Zone, everybody is friendly and obliging, to the extent that they will lie down in the road and let you run over them if you but ask. One of the inhabitants is having a rabbit problem. He has a carrot, and the rabbit is trying to steal it from him. He says ” Go away, little bunny! – I mean it!”. But the rabbit knows he doesn’t and keeps trying to bite it out of his hand.

      But when the old lady’s psychic power is suddenly extinguished, that rabbit is roasting on a spit in less than a minute. Rabbits are not inherently formidable.

    • athEIst says:

      There are a few more explosions. The Arab explosion of the 7th century into the Byzantine Empire, the Mongol explosion of the 13th century into the Caliphate and Russia, and the West European explosion of the 17th century into the western hemisphere. The last gave us the modern world, the second ultimately went nowhere, and the first gave us the present middle east.

    • Carl says:

      Economics does drive culture, and vice versa. It’s not all one thing or the other. There are dumb ways to arrange the economy, and these dumb ways can be (and have been) tried by smart people.

  10. Jim says:

    I suppose the “genetic diversity” among Askenazi Jews is a lot less than among the entire population of Sub-Saharan Africa. But the contributions of Askenazi to high IQ fields are simply stunning. A few months ago I made up a list of the top chess players of the last 150 years or so and then determined which of them were Ashkenazi. I expected a high percentage to be so but I was still surprised at the result – 49%.

  11. Dale says:

    McClosky might be right, but it’s not that important. Yeah, if all 750 million sub-Saharan Africans were raised in prosperous, industrialised circumstances and given good educations, and from them you filtered the best mathematicians, the top 1-in-100,000 in math ability would be quite good, and possibly better than you’d extract from other populations. But if you could actually do that will all the sub-Saharan Africans, then Africa’s problems would already have been solved; right now no more than a few tens of millions are raised with those advantages. Conversely, the ability to generate far, far outliers who are better than another culture’s far, far outliers doesn’t have much effect on the overall course of society, other than perhaps determining which culture generates the next Napoleon.

    Where all this gets interesting is when the cultural configuration of a population applies particular selective pressures on it for long enough to make a difference. What has been the impact on the aptitudes of the Chinese of living under hierarchical bureaucracies for 2,500 years? What are the consequences of Europe generating large states for the last 500 years?

    • gcochran9 says:

      You’re wrong. When you have a big different in the mean, the fraction that exceeds a high threshold becomes very small. If you look at how blacks in the United States do on standardized tests, they produce very few individuals that score at the highest levels – so much so that you can find single non-black families that can produce more such individuals than the entire black population of the United States. As I have mentioned before, the Bekele family, from Bekoji, Ethiopia would be likely to win a running competition against the entire Han.

      • Jim says:

        An extended family get together of the Novikov’s and their relatives would be pretty impressive.

      • Economic Sophisms says:

        This is such a huge point that people don’t get about normal densities. There’s essentially ZERO chance that a Sub Saharan population could produce a v. Neumann, hell that might be outside the realm of possibility of gentile Europeans or even E. Asians.

        • Space Ghost says:


          • Economic Sophisms says:

            certainly there’s a long list of Goyim who were ‘more important’ than v. Neumann but in terms of raw horse power, he sounds like the tops. Who knows though? At a certain point it becomes hard to tell the giants apart.

              • Jim says:

                Remember the clever evaluation of the Eulerian integral in calculus? The result was found by Euler but the clever argument found in every calculus textbook? That was Gauss.

              • Jim says:

                In the differential geometry of surfaces before Gauss the Gaussian curvature was hiding behind every formula but only Gauss saw it.

              • Carl says:

                Derb plumps for Gauss as genius par excellence, too

            • Bob says:

              Von Neumann’s “raw horse power” is a bit exaggerated. One of the big myths is that von Neumann was a great mental calculator — in fact he was not so at all, he simply memorized certain physical constants and algebraic techniques (there used to be a terrific bag of techniques back when the word “computer” referred to a human who calculated as a profession — look up “nomography” for example). This is because calculation is purely a matter of memorization and recall, and if von N. really was such a mental calculator then he could memorize text verbatim like Euler (who did have absolute recall) — and yet he had difficulty remembering people’s names. In terms of his proofs (e.g., in “Rings of Operators”, Parts I-IV), there is nothing at all to indicate any special faculty. There is a certain similarity in spirit between the way that von Neumann and Norbert Wiener thought. In their case, it has to do not with the originality of a certain construction, but of restricting the set of ideas in just the right manner, so that by taking combinations of them, the machinery (=formalism, language) itself would be powerful enough to smash open the problem. Wiener would, for example, translate any problem at all into a Wiener-Hopf problem, and then pull through by analytic tricks accumulated via memory. (A “baby example” of that — that I presume everyone here can follow — is the following simple problem involving random lengths: if a straight stick is cut up in three pieces at two points independently along its length, determine the probability that the three sticks can be made into the sides of a triangle [i.e., the sum of any two lengths must be larger than the third]. All you have to do, is to translate this into the language of (independent, continuous) random variables and uniform distributions — and then after everything had been completely translated, the problem simplifies almost to the solution via simple algebra.)

              Consequently their proofs are not elegant or efficient at all, even though once in a while it would be powerful enough to establish the first cracks in a problem. For example, G. D. Birkhoff, after reading von Neumann’s proof of the “mean” ergodic theorem, took just one week to prove his own (much better) version via “time averages”. Birkhoff’s proof was far more fundamental and difficult, than any single thing von Neumann did.

              • Jim says:

                What do you think of Von Neumann’s theory of ordinal numbers?

              • Jim says:

                I would have thought that Von Neumann’s theory of ordinals could be described as “elegent and efficent”. Admittedly one needs the Axiom of Foundation to carry it out.

              • Jim says:

                I looked through the index of names in Dieudonne’s History of Functional Analysis and found the following with 9 or more citations –

                                               Hilbert                    75
                                               Fourier                   47
                                               Riesz                     40
                                               Banach                  38
                                               Poincare                32
                                               Dirichlet                 29
                                               Fredholm               28
                                               Cauchy                  27
                                               Green                    27
                                               Von Neumann       25
                                                Schwarz               24
                                                Weyl                     21
                                                Liouville                18
                                                Frechet                 17
                                                C. Neumann         17
                                                Sturm                    15
                                                Schmidt                 15
                                                Volterra                 15
                                                Hahn                     14
                                                Lebesgue              13
                                                Poisson                 13
                                                Carleman               12
                                                Lagrange               12
                                                Hadamard              11
                                                Laplace                  10
                                                Toeplitz                  10
                                                Weber                    10
                                                Hellinger                10
                                                Helly                       10
                                                Weierstrass             9
                                                Gauss                      9
            • ckp says:


            • Cplusk says:

              Von Neumann, Terry Tao, Ed Witten, Einstein, Ramanujan. My top 5. (also Gödel)

        • RCB says:

          There is another fact about real numbers worth considering: for any arbitrarily small positive number, I can name another large positive number that makes the product of the two numbers arbitrarily large. (And there are a lot of East Asians.)

      • Rick says:

        If the Bekele family was smart, they would be selling aliquots of sperm to the highest bidder. And then later, after interest decreased, they would be selling them to anyone at a reduced price. That is how you create a genetic legacy in the internet age.

    • Erik Sieven says:

      i think 750 million was a few years ago, by now it is over 1 billion.

    • pyrrhus says:

      With the average IQ in subsaharan Africa being around 70, that suggests that only around 1 in 36 would have IQs of 100. Roughly speaking, that means that the higher levels of intelligence would be about 1/40th as frequent in African populations as in European populations… everything about your concept is wrong.

      • Ursiform says:

        The higher you go the larger the ratio becomes.

        • Ursiform says:

          Assuming a standard deviation of 15, a population with mean IQ of 100 will have about four people per thousand with an IQ over 140. With a mean of 70 it’s about one and a half per million. Assuming a mean of 70 and std of 15 there would be about five black Africans with an IQ over 150.

        • RCB says:

          One thing that no one seems to talk about is that the Gaussian’s variance really dominates for very distant deviations.

          Suppose that Population A has mean 0 and SD 1, and Population B has mean -1 and SD 1.3. Then Pop A will have about twice as many individuals exceeding 2. But Pop B will have twice as many exceeding 4. And 37 times as many exceeding 5. If we were talking about IQ, then Pop B would be dumber on average, but produce almost all of the once-in-a-generation geniuses. (Assumed both pops are same size.)

          Of course, once you get really far into the tails, the Gaussian is no longer a good approximation of reality. We don’t actually have an infinite number of loci, after all.

          I don’t know what the variance in IQ is for black Americans. Not that it matters: we already know that black Americans don’t produce more geniuses than do white folk.

            • Supporting Greg’s remark that African American intelligence probably has a sd of about 14. As usual, would be good to look at most recent data to see if there have been any changes in that general picture. By the way, the most recent yet to be published work on sex differences (n about 900, all MRI scanned) shows 4 point male advantage, of which about 2 points probably due to bigger brains. Will post about it when it is in the publication pipeline.

      • Jim says:

        Lynn estimates that Sub-Saharan African IQ under optimal conditions is probably more like 80 than 70. The average IQ in the US Virgin Islands is 78. As this is a US Territory I assume conditions are pretty decent – clean water, food stamps etc. so Lynn’s estimate is probably not far off.

        At this level about 9% of the population would have IQ’s above 100.

  12. RCB says:

    Re Ashraf and Galor

    I recall this paper coming out in grad school. The paper is pretty dumb. Basically their argument (someone correct me if I’m not remembering it correctly) is that if your nation’s genetic diversity is too low, you don’t have enough variation in skill, or something, and so you can’t get the variety of skill you need for specialization. But if your genetic diversity is too high, then everyone is too genetically different, and so there is no cooperation, because “kin selection”. Somewhere in the middle is the sweet spot. Where, exactly? Wherever they happen to find it, of course. (Of course, given the approximate parabolic curve in the data, any two competing hypotheses would have fit just as well.)

    Cochran has already explained why the diversity bit is mostly nonsense. Turns out the kin selection bit is also nonsense, for a variety of reasons: e.g., heterozygosity does not enter the equations of inclusive fitness; most reproductive competition is within, not between, nations. But inclusive fitness is a tricky subject; amateurs mess it up all the time. Frank Salter’s ethnic genetic interests is another example of messing it up.

    But the funny part about the paper was the reaction to it. Even in my evolutionary anthropology department (UC Davis), talking about race differences in any trait of practical importance was taboo. So people “knew” it was wrong on principle, and yet most didn’t know enough about genetics to explain why. We weren’t the only department: a group of Harvard anthropologists quickly responded with the obligatory many-authored, one-page, moralistic critique ( Of course these kind of responses are so trite that they virtually write themselves. Maybe I should write a program that scans google scholar for new scientific papers with racist implications, writes a quick condemnation, and submits it to the same journal. I’d be a hero.

    • Carl says:

      The problem with these taboos is that genuinely racist i.e wilfully malicious people will be among the first to break them, and not in the noble spirit of science. Doesn’t have any bearing on the facts of the matter, obviously, but it entrenches the taboo further. If I see another neo-Nazi citing average group differences before recommending genocide or compulsory sterilization I will despair. We need more Nicholas Wade or Steven Pinker types to encourage others. Breaking the taboo needs to become associated with high status, somehow. It has happened before. This will infuriate those who resent the idea of having to soft-peddle the truth or spray “squid ink”, as Derb puts it.

  13. teageegeepea says:

    When just McCloskey completely bungled the concept of “regression to the mean” while reviewing “A Farewell to Alms”, that was one thing. When Sam Bowles (who should know better) did the same, the profession as a whole starts to resemble your critique of it.

    • gcochran9 says:

      Where does Bowles do this?

      • RCB says:

        Clark seems to think that Bowles makes the mistake here:
        But I don’t see any explicit mention of regression to mean here. He does suggest that some personality traits apparently have very low heritability.

        The difficulty with regression to the mean is that arises from two mechanisms: imperfect heritability and imperfect assortative mating. I don’t think people understand that the dynamics of these two are different. The former is really a one-generation effect; it doesn’t compound over time. The latter does. So if you have moderate heritability and high assortative mating, you can definitely get long-term persistence of traits in particular families, classes, etc.

        I had the pleasure of talking with Greg Clark in a long van trip to Santa Fe Institute once. He understood the difference between these two mechanisms, despite (I suspect) little formal training in genetics. That’s because he is very smart.

    • RCB says:

      Here’s a quiz I like to give to see if people understand regression to the mean. I suspect readers here would do pretty well.

      (1) Two smart parents have a mean measured IQ of, say, 120. Do we expect their son to have a IQ lower, higher, or about equal to this?
      (2) A son’s IQ is measured to be 120. Do we expect the parents’ mean IQ to be lower, higher, or about equal?
      (3) A particular asexual organism (call it Pat) has a quantitative trait deviation (difference from the mean) of, say, 1. Pat has many offspring, who altogether have a mean deviation of 0.9. What is the expected mean deviation of their offspring, i.e. Pat’s grandchildren?

      • reiner Tor says:

        My guess:

        (1) lower
        (2) lower
        (3) 0.9

        • reiner Tor says:

          My reasoning:

          (1) a 120 IQ person might have some luck besides the good genes – either negative (in which case his offspring will inherit smarter genes) or positive (in which case the offspring will inherit dumber genes). Unfortunately, the number of people genetically endowed with >120 IQ is smaller than those <120 IQ, so it’s more likely his luck was positive. Since this is true for both 120 IQ parents, and their offspring could get luck either way (or no luck at all), the most likely explanation is that the offspring will be <120 IQ.

          (2) Same thing, the 120 IQ son is more likely to be somewhat lucky than unlucky, so his parents are probably <120 IQ.

          (3) Pat and its offspring are genetic clones, so the fact that the many offspring on average have 0.9, whereas Pat alone is 1.0, suggests that Pat has some luck. It’s unlikely that Pat’s grand-offspring will have (on average) luck either way, so 0.9.

      • 1) lower. Say 112 if heritabilility taken as .6
        2) lower. More parents are below 120 than above.
        3) Still thinking about it.

      • ckp says:

        1) lower
        2) not sure, see below
        3) 0.9

        If the son’s IQ were 120, we might reason that it had regressed from a higher parental mean. However, the probability of a high parental mean is a priori lower. I can’t figure out which effect dominates in my head, I’d need to calculate it.

      • tautology123 says:

        1) lower
        3) (0.9*n + 1)/(n+1) where n is the number of offspring. If n is sufficiently large, this is close to 0.9.

      • RCB says:

        Y’all done well. Lower, lower, 0.9. (tautology123 went the extra mile by noting that the parent is also a sample from the genetic type, but really when I said “many offspring” the lazy implication was “infinity”, so 0.9 is fine.)

        Now if you really want to waste time, here’s a bonus question: given genetic and environmental variances Vg and Ve, provide the posterior probability distributions for the son’s phenotype (in problem 1) and the midparents’ phenotype (in problem 2). Assume segregation variance is 1/2*V_G. Both posteriors will be Gaussian, so you can just provide mu and sigma. (mu tells you the best estimate.)

        • tautology123 says:

          Ok, not sure about that one.

          1) mu=20 Vg/(Vg+Ve) (simply response to selection)
          (Environmental variance will affect the offspring in every case. Additive genetic variance is fixed, due to parents being fixed, but segregation variance is an additional component ?)

          2) mu=20 Vg/(Vg+Ve) (response to selection again, as it is time direction invariant)
          (environmental variance acts in any case and segreagation variance is again symmetric, but only coming from half the genome of the offspeing. However each midparent does not only have half the genes of the offspring, but also another half of genes again having additional genetic variace. Since it is half a genome this variance is Vg/2 ?)

          Tell me how I am wrong.

          • RCB says:

            Generalizing, let Xp be the midparent phenotype and 0 be the mean phenotype. Assuming random mating, then, in question 1,
            mu = Xp Vg/(Vg+Ve) (so you got this right)
            sigma^2 = 1/2 VgVe/(Vg + Ve) + 1/2 Vg + Ve (you were close)

            You were close, but missed 1 source of variation. Think of it this way: To get from midparent phenotype to offspring phenotype, we first move (backward) to midparent genotype (the term you missed), then to offspring genotype, and finally to offspring phenotype. Each of these steps adds noise; they are the three variance terms above, respectively.

            The exact form of the first term is not obvious at all (I didn’t reason it; I derived it). But it makes some sense. If Ve=0, then parent phenotype = parent genotype, so there is no noise added. If Vg=0, then the parent genotype must be 0 (the mean), so no noise added there either. So it makes sense that the term would be a product of both.

            I am too lazy to do the math for the 2nd problem, but I suspect the variance term will be even more complicated; I doubt your answer is quite right.

  14. Economic Sophisms says:

    Economics is bloated and full of itself. It’s good to have some professors to teach students, and it’s good to have a few researchers so we can get important historical works and better understand big things like monetary policy, but otherwise it’s a pretty useless profession. Here’s a challenge: go to the home pages of 5 random tenured econ professors at the research college closest to you. See what they’ve published on, and ask “does this matter at all?”. The world would be better off if a lot of these people were working in the private sector, or some useful applied science field in academia; waste of good brains.

  15. “Europeans have less overall genetic diversity than sub Saharan Africans (~20% less)”

    I find this interesting. Why so little? One would assume that since Europeans came from a bit of a population bottleneck (I’ve been told) 70,000 years ago that there would be a higher percentage ofgenetic diversity than 20% when comparing the gene pool of Europeans and sub Saharan Africans

    My question is why isn’t that percentage higher? Has recent evolution (new alleles) of the vastly larger population that left Africa accounted for considerable genetic diversity in Europeans?

    • gcochran9 says:

      ~20% less means 80% as much.

      • I got that, let me rephrase what I am saying. One of these is wrong.

        1) A small population left Africa and populated the rest of the world.

        2)There is 80% as much variation between Europeans and sub Saharan Africans.

        I’m guessing #1 is wrong because human dispersal has been a braided stream far more than the “out of Africa,” but I don’t know.

        • gcochran9 says:

          #1, roughly speaking. I’m not even sure what you mean by #2.

          • Frank says:

            Roughly speaking is correct. If you are comparing only small local populations, the difference is much bigger.

            Even a very small khoisan group will have more heterozygosity in the average individual than is found in many large Eurasian populations because they have had the largest effective population size for the longest amount of time.

  16. pyrrhus says:

    You are being too polite, Dr. Cochran. Deirdre McCloskey’s book is just the usual SJW academic rubbish based on wishful thinking and a strong aversion to reality. I wouldn’t be surprised to see it win awards….

    • SamGamgee says:

      McCloskey is hardly SJW material. What SJW do you know of praises bourgeois capitalism and the virtues of free enterprise? No, she fits into that particular type that Steve Sailer has identified: high-achieving and quite assertive and masculine, with little sympathy for feminizing doctrines of equality and social justice, but erotically fascinated with identifying as a woman, which in our confused PC culture gets conflated with more genuine transgendered individuals, i.e. the extremely shy and feminine types that are easy targets for bullies and for whom gender reassignment is often the only alternative to suicide. Along with Deirdre McCloskey, you can include Caitlyn Jenner.

      • It’s called “autogynephilia,” and it got shouted down without a hearing at the APA convention a decade ago, but it’s quite real.

      • Greying Wanderer says:

        “What SJW do you know of praises bourgeois capitalism and the virtues of free enterprise?”

        Wrong way round: what globalist pro-capitalism type people do you know who are also now publicly SJW?

        Most of them.

        The cheap labor lobby have taken to using SJW open borders arguments (blank slate being the biggest) because the economic arguments they used in the past no longer stand up to reality.

        • Carl says:

          Well, if we’re going to call McCloskey a SJW we need a new term for the actual SJWs.

          • Greying Wanderer says:

            I’d keep SJW as is but just recognize globalists have adopted fake SJWism cos their economic arguments have been proved wrong.

  17. (Greg): “In a century or so the leading scientists and artists in the world will be black – at any rate if the diversity is as large in gene expression and social relevance as it is in, say, height or running ability. ”
    The conditional “if” matters to the argument.
    (Greg): “McCloskey is thinking that a turn to free markets will make Africa rich, and that will give educational opportunities to Africans now denied them – but a fair-sized mostly-African population already lives in the United States, a population that is already much more prosperous than sub-Saharan Africans. How are they doing? How many geniuses are they producing?”
    This does not rebut McCloskey’s argument. The American black population descended from a narrow sample of the African population. Most of the variation remained in Africa.

    • gcochran9 says:

      Any human population has most of the variation found in the whole human race. West-African types made up most of the slaves brought to the US, along with some from farther south (Angola) and a smaller fraction from Mozambique. Probably very few were Bushmen or Pygmies. Since most of the population of sub-Saharan Africa is a product of the Bantu expansion, which originated in west Africa, US blacks are not that narrow a sample of African diversity, not that diversity matters anyhow.

  18. SamGamgee says:

    It just sounds like McCloskey wants to eat her cake and have it, too. She reminds me a bit of Jared Diamond and those genius Papuans, who was happy to entertain the idea that some ethnic groups were genuinely more intelligent than others, as long as blacks were smarter. So McCloskey seems eager to accept that genetic differences can lead to different economic outcomes, provided that the worst-performing group turns out to be the best.

    • martin says:

      It used to be taboo to discuss any racial differences whatsoever, but now the situation has improved so that one can mention racial differences in polite company, providing that the racial differences in question are to the detriment of white people.

    • TWS says:

      I think you forget the aerospace program of the deep interior of Papua New Guinea. They employed a significant portion of their population and built many, many airplanes. I don’t know how many made it off the ground but they sure built them.

    • ursiform says:

      I think Diamond is a pretty smart guy lead astray by his penis sheath envy …

    • Yudi says:

      It’s the racial version of the Althouse rule.

  19. Pingback: “See No Genes, Hear No Genes, Say No Genes”–Greg Cochran Refutes Deirdre McCloskey | VDARE - premier news outlet for patriotic immigration reform

  20. MawBTS says:

    If a population exists at the top of some quantitative trait, wouldn’t you expect LESS less diversity by default?

    A die with six on every face is going to be pretty unvaried in how it rolls. Change any of the faces and you’ll get 1) more diverse rolls. and 2) a worse-performing die, every time. There’s nowhere to go but down.

    (Assuming that six is a hard limit, and that you can’t have a die that rolls a seven.)

  21. tommy says:

    It’s high time somebody wrote the definitive guide to fallacies in genetics and biology, complete with superb analogies and real-life examples, for all those outside the field who would rely upon its evidence for their arguments.

    • ckp says:

      Counterpoint: denialists will produce their own definitive guide in response. The layperson will find it difficult to decide between them because the two competing theories seem equally plausible in the absence of domain knowledge.

  22. Neocolonial says:

    Off topic: How does something like the difference in throwing action consistently observed between men and women develop / be transferred? And even between a man’s good and off hands?

    • Sideways says:

      Try throwing with your other hand. You’ll throw like a girl.

      Just a matter of training and practice.

      • Frank says:

        This is not just training and practice.

        The human brain is usually set up to prefer one side over the other for certain tasks.

        Back when they used to force kids to be right handed, the ‘naturally’ left handed never wrote as clearly as the others when using their right hand.

        Perhaps that had to do with having a minority-sided (some might say ‘wrong-sided’) brain to begin with.

      • DrBill says:

        “Just a matter of training and practice.”

        Does that really seem likely to you? Do boys start out throwing like girls and then learn better? By trial and error? I doubt. Here is an article which claims that Abo females, who it claims throw as much as do Abo males, also throw like girls.

  23. syonredux says:

    ” Is there evidence, either in test scores or cultural accomplishment (better than any test), that some African populations may have smarts comparable with, or better than, people in Switzerland or Holland or Scotland?

    No. There is no such evidence.”

    Chanda Chisala has been trying to persuade everyone that the Igbo are the Sub-Saharan version of the Ashkenazi. Of course, they still haven’t managed to produce any Einsteins….

    • gcochran9 says:

      In principle, the IQ selection fairy could landed upon the heads of some population in Africa (or anywhere) – but if it had we’d know already. Unless they were hidden in a lost valley – six adventure novels follow.

      My first calculus instructor in college was igbo. he was a good guy, competent: I liked him. N = 1.

      Now things may change. I like the idea that genetic enhancement of IQ will become possible, but illegal and/or hazardous – so some desperate oppressed group takes the leap and astounds us all.

  24. Rikard says:


    A wild speculation: the percentage of Neanderthal DNA is a contributing factor to the difference between sub-Saharan africans and europeans.

    Another wild speculation: reproduction/carnivore-strategy difference has led to europeans being more succesful in using the intelligence they’ve got compared to the more reproduction-strategy oriented africans.

    Since I’m neither a geneticist nor an economist I am not making any claims to validity of argument, but perhaps the ideas are interesting.

    Comradely greetings from Sweden,
    Rikard, schoolteacher

    • Greying Wanderer says:

      I don’t know if the Neanderthal percentage matters as I’d have thought any critical genes would take up the same (small?) percentage(?). The percentage might be more an indicator of time since admixture or a slower rate of decline and include things that don’t matter much in the modern age – like size, facial phenotype etc.

      However people who needed to get 12 months food in 8 months would need a different time preference to people who could get food all year round – and as time preference acts as a magnifier for almost everything that is important in the modern world it seems like it would be critical.

      So personally I think some Neanderthal admixture is/was critical but in the modern world not so much the percentage – although the percentage may have been more of a factor in the past’s sword-based meritocracy.

  25. IC says:

    If you are good at math, this puzzle is not hard to solve.

    If genius is like genetic lottery combination number, a limited pool of most available numbers concentrated on those winning numbers yield more winning lotteries than diverse pool with all possible numbers to draw from. It is the game of odd.

    Breeding of domesticated animals is a process of reducing such gene pool to create better odd. Human living in very high selection pressure will have similar result.

    With increasing knowledge about specific genes or alleles and genetic architectures contribution to intelligence, plus CRISPR tech, this game of nature lottery would end soon since most people would not like to the stupid ones.

    • Greying Wanderer says:

      “With increasing knowledge about specific genes or alleles and genetic architectures contribution to intelligence, plus CRISPR tech, this game of nature lottery would end soon since most people would not like to the stupid ones.”

      That’s the million dollar question.

      Would the current elite prefer a small high IQ elite with a vast mass of low IQ slave-cattle or would they dramatically increase average IQ partly by increase and partly by decreasing the variance at the low end?

      • Greying Wanderer says:

        If you take the West’s current immigration policy as a clue I think the answer in the West is option 1.

        If Russia, China or India take the second option we (or at least people in the future) will see which of the two options: smart fraction or high average, is the best.

  26. simontmn says:

    “Is there evidence, either in test scores or cultural accomplishment (better than any test), that some African populations may have smarts comparable with, or better than, people in Switzerland or Holland or Scotland?”

    Igbo – the women at least – seem to be comparable to Europeans, though maybe not to Germanics specifically.

  27. Montezuma says:

    When it comes to black achievement, people have to resort to lies:

    “Euclid and other African mathematicians outshone Europe’s brightest stars for millennia.'”

    ’nuff said.

    • RCB says:

      Says Farley, author of this article: “John Derbyshire, a columnist for the National Review, wrote an essay last week implying that black people were intellectually inferior to white people: ‘Only one out of six blacks is smarter than the average white.’ Derbyshire pulled these figures from a region near his large intestine.”

      Amazing. This is the first paragraph of the paper. And yet it’s immediately seen to be bullshit, even if you know nothing about the topic: Derbyshire didn’t pull these stats out of his ass; the very text of the original claim is hyperlinked to the Roth et al 2001 meta analysis – peer-reviewed, for those who care.

      How does this happen?

      • RCB says:

        On second thought, it’s likely that the Roth et al paper doesn’t explicitly spell out that 1 sd difference implies that only 1/6 of blacks will exceed the white mean under a Gaussian distribution. So, while Farley may have been smart enough to know that blue text = hyperlink, he probably wasn’t smart enough to google the cumulative distribution function of a Gaussian.

      • Julian says:

        How does this happen?

        Sacred values. Jon Haidt explains:

        The fields of psychology, sociology and anthropology have long attracted liberals, but they became more exclusive after the 1960s, according to Dr. Haidt. “The fight for civil rights and against racism became the sacred cause unifying the left throughout American society, and within the academy,” he said, arguing that this shared morality both “binds and blinds.”

        “If a group circles around sacred values, they will evolve into a tribal-moral community,” he said. “They’ll embrace science whenever it supports their sacred values, but they’ll ditch it or distort it as soon as it threatens a sacred value.” It’s easy for social scientists to observe this process in other communities, like the fundamentalist Christians who embrace “intelligent design” while rejecting Darwinism. But academics can be selective, too, as Daniel Patrick Moynihan found in 1965 when he warned about the rise of unmarried parenthood and welfare dependency among blacks — violating the taboo against criticizing victims of racism.

    • syonredux says:

      “The second story involves one of the few black mathematicians whom white mathematicians acknowledge as great – or, I should say, “black American mathematicians”, since obviously Euclid, Eratosthenes and other African mathematicians outshone Europe’s brightest stars for millennia”

      Let’s see, Eratosthenes was born in the Greek colony of Cyrene, and Euclid was from the Greek colony of Alexandria.So, they were about as African as Kipling was Indian.

      • Jim says:

        Alexandria was a great center of learning and attracted scholars from all over the Greek world. We don’t know where Euclid was born although there is no reason to question the tradition that his adult life was mostly lived in Alexandria. Euclid’s works were written in a technical jargon which does not allow his native dialect of Greek to be determined.

      • Jim says:

        It is also difficult to evaluate how much of what is in Euclid is actually original. His work wasn’t the first axiomatic account of geometry. Hippocrates of Chios is credited with having written the first axiomatic account of geometry more than a hundred years before Euclid. The works of Aristotle and Plato mention several other authors of axiomatic accounts of geometry. However in the Timaeus Plato complains of the neglect of three dimensional geometry. Euclid has quite a bit of coverage of three dimensional geometry so it may well be in that regard that his elements was superior to those of his predecessors.

  28. LL says:

    Hold up everybody. We’ve now found an area of intellectual achievement where Africans DOMINATE.


    Please adjust all worldviews accordingly.

  29. Pingback: Outside in - Involvements with reality » Blog Archive » Chaos Patch (#115)

  30. Dears,

    This blog seems to veer into into adolescent sneering pretty quickly, but let me venture a reply. I’m very willing to be corrected about the genetics, about which I am woefully ignorant. True, I did check my crude idea with a couple of human geneticists. They seemed to think it possible that the cross-African variation in genetic material would, when Africans had the same opportunities as rich countries, result in notable outliers, and indeed outlier populations, such as Kenyan runners. Some of the correspondents here do not seem to understand, as I explained in the book, that that plants and animals that migrate in small groups will lose genetic variation, from the small sample size of the migrants and the chances of not mating.

    Still, the hypothesis in question is one page in an 800-page book, the rest of which offers an explanation of modern economic growth having nothing to do with genetics. Someone whose intellectual life consists of making snarky comments on blogs without reading books will want to know that.

    On the unusual mathematical ability among Eastern European Jews there is a story. John Von Neumann, “Johnny,” was a witty and sophisticated man, and one of the great mathematicians of the last century. If a gentile happened to prove an important theorem, Johnny would rush into the seminar room at the Institute and breathlessly declare, “The goyim have the following result!”

    In peace, for the good of science,

    Deirdre McCloskey

    • MEH 0910 says:

      “Still, the hypothesis in question is one page in an 800-page book”

      It’s clear that Greg is passing judgment on this one page, not the whole 800 page book.

      To quote Greg:
      “Deirdre McCloskey has a new book out: Bourgeois Equality: How Ideas, Not Capital or Institutions, Enriched the World. I’m sure that there are many good things in it. But McCloskey makes a significant error in talking about genetics.”

    • Thras says:

      “Dears”…”adolescent sneering”…”In peace for the good of science” —

      I hope that the entire book isn’t written like that. I think that if I were going to go to the effort of dressing up and cutting off my junk, I would at least learn to write like a female. The above phrasing just oozes creepy maleness. It reads like someone LARPing as a girl in an online game, who is mostly just sexually turned on by imagining himself as a woman. Gross.

      Spend some time reading some real world women to pattern your writing after them before transitioning, folks! Otherwise people are just going to imagine that you are just a poor sob who broke his brain through too much masturbation while wearing female undergarments.

      • This is intellectual debate? No. Adolescent sneering.

        • gcochran9 says:

          You know nothing about genetics, except for a few things that aren’t so. Therefore, when you opine on it, you are reliably wrong. Same thing happened when you said that a selected population keeps regressing all the way back to the mean, when you were arguing with Greg Clark – that’s wrong. If that notion were true, natural selection would be impossible, and neither Guernsey cows nor Chihuahuas could exist.

          Many people feel that they are entitled to have opinions – ones that should be taken seriously – on subjects of which they know nothing. Oddly, they keep picking the same subjects, like genetics – hardly ever electrical engineering or number theory. Do you have opinions about the ABC conjecture, or on twin primes? I didn’t think so. Just as well.

  31. Fst says:

    “Bushmen are genetically more distant from the Bantu than the Bantu are from Chinese.”

    That can’t be true. Fst distance for Bantu-S.Chinese is 2963, and for Bantu-San it’s only 94.

  32. Check it out. Her debate with a philosopher of economics on whether the standard correspondent theory of truth is correct(she is against it).

    Click to access Article_164.pdf

  33. Scott Locklin says:

    Isn’t McCloskey’s the “economist” who cut his own dong off? Weren’t you just saying that we have an insane overclass in the previous blog, or am I reading things backwards?
    I am assuming you’re trolling us.

  34. Pingback: This Week in Reaction (2016/05/22) - Social Matter

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 )

Connecting to %s