Is poverty in our genes?

Is Poverty in Our Genes? is the title of a new extended critique of Ashraf and Galor’s forthcoming American Economic Review paper on genetic diversity and economic development. Published in Current Anthropology, the critique is an extension of an earlier piece by a group of academics (mainly from Harvard) who argue that Ashraf and Galor’s work is false and undesirable.

The critique spends some time focusing on the data underlying Ashraf and Galor’s work, which provides a good reminder of the complexity of human migratory history. For example, the authors write:

Historical flaws also exist in Ashraf and Galor’s treatment of concepts of innovation in table A3. Here the achievements of the diverse populations at Cordoba are taken to stand for measures of “European” innovation at 1000 CE. It is misleading to use Cordoba as a measure of European success, given that it was ruled by North African Moors until 1236 CE. Likewise, it seems inconsistent to classify Constantinople as part of Europe in 1000 CE but part of Asia in 1500 CE (Ashraf and Galor 2013, table A3). It should also be remembered that Europe’s role in innovation is a very recent phenomenon. Indeed, if we are to look for traces of “innovation” according to Ashraf and Galor’s standards in Europe, archaeology has made it clear that agriculture was not independently invented in Europe, but rather spread there from the Near East (Bellwood 2006). One can also show that Renaissance Europe was heavily influenced by Greek and Arab thought (Lewis 2009; Saliba 2007). Clearly, there is a great deal of multicontinental interaction in the circum-Mediterranean region. If one excluded these data coming from the heavily African- and Middle Eastern–influenced Mediterranean region, population levels (and hence innovation levels, according to Ashraf and Galor) in Europe would be low compared to other areas of the world until the late medieval period (after 1470).

These are interesting arguments, but I’m not convinced that shifting a few data points will materially change the general findings. The more fruitful area of criticism is the causative mechanism. In that area, the authors make some interesting points about evidence from other species.

Ashraf and Galor’s theoretical model argues that genetic diversity can play a positive role in the expansion of a society’s “production possibility frontier” or its ability to innovate. In their appendix H, they use animal studies to justify this claim. They describe studies on insects that link genetic diversity to disease resistance and to several aspects of hive performance in honeybees (Seeley and Tarpy 2007; Tarpy 2003). The two bee studies cited by Ashraf and Galor correlate genetic diversity with bee foraging rates and hive temperature and indicate that disease susceptibility relates to inbreeding. Another cited insect study on fruit flies (Drosphila species) shows that genetic diversity helps increase resistance to environmental changes (Frankham et al. 1999). It is unclear how either of these relates to an ability to innovate. Perhaps Ashraf and Galor were inspired to use these data because there is no research demonstrating that genetic heterozygosity at the population level is associated with capacity to innovate.

In addition, these cross-species comparisons of genetic diversity seem to not take into account how genetic diversity varies widely among species. Humans are noted for having extremely low levels of genetic diversity compared to other animals, including our closest cousins, chimpanzees. In fact, some chimpanzee breeding groups, such as those in the Taï forests of West Africa, are estimated to have greater nucleotide diversity than the entire human species (Gagneux et al. 1999). It is important to put into perspective that the total amount of human genetic diversity is actually quite small compared to that found in other model organisms.

My posts on Ashraf and Galor’s paper on genetic diversity and economic growth are as follows:

  1. A summary of the paper methodology and findings
  2. Does genetic diversity increase innovation?
  3. Does genetic diversity increase conflict?
  4. Is genetic diversity a proxy for phenotypic diversity?
  5. Is population density a good measure of technological progress?
  6. What are the policy implications of the effects of genetic diversity on economic development?
  7. Should this paper have been published?

Other debate on this paper can also be found hereherehere and here.

Comments

  1. Dr. Fine says

    The extended 4-page comment by the anthropologists is as shallow as their earlier letter that was ridiculed by Ashraf and Galor.

    I am eager to read the next response of AG, but I cannot resist the temptation to reveal myself some the shallowness of these ill trained anthropologists.

    The main substantive point raised in the comment of the anthropologists is that the population density data of MacEvedy and Jones is imperfect for the American continent in the pre-1500 period. (A criticism that, as one should expect by now from these anthropologists, is based on few anecdotal evidence). Is this a source of concern? Absolutely not.

    1. Ashraf and Galor show that the hump shape effect of genetic diversity on economic development is present in the year 2000, when income per capita is used. How is the argument about population density in 1500 relevant for this main finding?

    2. Even if there is a systematic measurement error that affects that data from the Americas in 1500, as the anthropologists argue, Ashraf and Galor’s inclusion of continental fixed-effects would account for it (see, http://ideas.repec.org/p/bro/econwp/2010-7.html Column (6) of Table 3). That is, AG identify the effect based on variations within a continent, not across continents. I realize that the cultural anthropologists have limited training in statistics, but I would hope that someone in their academic community could explain this trivial point to them.

  2. Passerby says

    It is interesting to note that a list of authors of the critique has changed – quite a few notable anthropologists dropped out and did not put their names on the piece published in Current anthropology… It seems that the piece in CA is authored mostly by grad students and young assistant professors.

    • Christina Warinner says

      If you check the author list you will see that is not true. Six authors are full professors, 2 authors are associate professors, 2 authors are assistant professors, 2 authors are senior researchers, 2 authors are postdocs, and 4 authors are graduate students. Please don’t post false information.

      • Passerby says

        If I check the author list, I see that it is true. 44% of the authors are assistant profs or below. With all due respect, in a paper signed by 18 people it is fairly clear who does the writing. So, yes, it is mostly authored by young researchers. Their passion is evident and is commendable; less commendable is their lack of experience and shallow arguments.

        More importantly, 39% of original signees dropped out (7 out 18).

      • passerby says

        I should also point out that the CA piece has never been peer reviewed (as far as publishing info indicates – it was submited on 24.10.12 and accepted on 25.10.12). As such, it is not much different from an op-ed piece or a blog post.

  3. C.W. says

    I’m in the discussion of the Ashraf and Galor paper for some month now and never made any comments online to that but now I feel that it is time to make some remarks to the nw critique of the anthropologists:

    1. The critique of the use of the McEvedy and Jones population density data is (as already noticed by the first comment) not reasonable.
    McEvedy and Jones (1978) is the standard source for cross-country historical population estimates used in dozen of papers by economicst, historians and economic historians. No Referee – as I now- did refuse this data source or demaned authors to use another. Of course – as with every historical source- the uncertainty is relatively high and increases the farther backwards you go – this is because Ashraf and Galor use that of 1500 AD and put that of the other years in the appendix. Of course, it might be the case that the figures are not correct for some periods or regions. Does this make the source completely unreliable and useless? No, because there always will be measurement error – especially in historcial measures. Additionally, Acemoglu et al. (2002) did conduct their empirical estimates based on different population estimates (like that of Bairoch etc.) and found no significant differences at all – at least for their regressions.

    2. Pretty much the same holds true for the “timing of agricultural transition” variable from Putterman (2008) . This is already a rough measure (no body would argue that it is possible to figure out the exact date of the agricultural transition) and it might be uncorrect for some countries. But, are the measurement erros systematic and how large is this error? As long as these questions cannot be answered appropriately, one should not dismiss this measure.
    Of course, I know that actually the measure is critized also by some economists (Acemoglu and Robinson e.g. in their new book and blog “Why nations fail”) and I think, there will be a new measure in future. But today the Putterman (2008) measure – already used in several papers – is the best and far most comprehensive collection of “timing of the Neolithic revolution” estimates available (as far as I know). Therefore it was an understandable choice to use this measure.

    3. Again, the same is the case for the trust question from the WVS. This is actually the standard measure for generalized trust (i.e. general and unspecific trust of people in other people or strangers) used in almost every empirical study on trust by economist, sociologist or political scientists in the last 20 years.
    It was – and is- heavily critized by economist and others. As reaction, a significant amount of experimental economic papers test the validity of these question in field or laboratory experiments. Although no clear picture did emerge from these studies (in my opinion) the majority of the papers conclude that the question is valid , i.e. is correlated with acutal trusting behavior in experiments. Furthermore, the question clearly is related to the amount of returned wallets in a so called “lost wallet game”.
    Of course, the measure actually is not perfect and there are still economist who do not believe in studies based on this question from WVS. Nevertheless, it is the (only) standard measure established in the literature.

    4. The author of the critique claim that distance to Addis Abeba would only be a proxy for genetic diversity on a contintal scale and does only proxy broad global trends. They wrote:

    “Ashraf and Galor’s description of the human pattern of global
    genetic diversity is consistently inaccurate, leading to concerns
    that the authors do not understand the data they are attempting to characterize. For example, they repeatedly contend that “migratory distance” to various settlements across
    the globe affected genetic diversity. This is misleading. The
    pattern of human genetic diversity they are referring to was
    primarily affected by the sequential series of founder effects
    that occurred during the peopling of the world; geographic
    distance is largely a proxy for these founder effects (Ramachandran et al. 2005). This proxy is accurate for roughly
    predicting global trends of genetic diversity on a continental
    scale but does not predict regional genetic diversity within
    continents. Human populations, stratified by heterozygosity,
    can be grouped into just four classes: Africa, West Eurasia,
    East Eurasia, and a fourth class comprising the remaining
    populations, nearly all of which have low heterozygosity. This
    class includes Native American populations. We prefer to use
    sequence data rather than genotype data to measure heterozygosity, as this avoids ascertainment issues involving the
    choice of SNPs used. Table S36 of Meyer et al. (2012), which
    used high coverage sequence data from 11 humans, shows
    the pattern clearly. In other words, genetic diversity varies on
    a continental scale, with Africa the most diverse, the Americas
    the least, and Eurasia having intermediate values. No amount
    of regression analysis and bootstrapping can alter the fact that,
    in essence, Ashraf and Galor are working with only four data
    points: Africa, Europe, Asia, and the Americas. This would
    be the case even if the raw data of Ashraf and Galor were
    perfect and free of noise.”

    I read the paper of Ramachandran et al. (2005) and if one sticks to that paper, this is not true ( I do not know how accepted and established the findings and methodology of Ramachandran et al. (2005) are within population genetics, but at least Ashraf and Galor stick primarly to them). Ashraf and Galor do extensively discuss and explain the findings and methodology of Ramachandran et al. (2005) and then – in my impression- do exactly the same as Ramachandran et al. (2005) . They regress the genetic diversity measure on the distance to Ethiophia taking the migration patterns of humanity into account. This corresponds exactely to what Ramachandran et al. (2005) (Figure 4A) do. Additionally, they found evidence for a clear relation between pairwise genetic distances between the ethnic groups in their sample and the great-circle distance (with waypoints and without) between them (Figure 1 in Ramachandran et al. 2005). So, obviously, at least according to Ramachandran et al. the distance and diversity relation does not only hold for continents or broad global tendencies. Or do I get that wrong (I’m economist not anthropologist or biologist)
    I think the only point where Ashraf and Galor can be critized here, is that they do rely on only one paper and do not discuss the probably existing other work in the area. But I do not know whether the other literature come to other conclusions than Ramachandran et al. (2005).

    Nevertheless, I think the anthropologists do have some good points.
    First, the fact that genetic diversity between humans in general is remarkably low. This is a huge problem for all arguments highlighting that genetic diversity between humans might be important for xx or can explain this or that.

    Second and probably most important, they correctly critize that the causal story of the paper is really, really weak. To make causal claims about very general and unspecific advantages and disadvantages of genetic diversity within(!) populations and genetic diversity associated with diversity in “Junk DNA” is not very convincing. And then, even worse I do not see enough convincing arguments to link both diversity in “Junk DNA” and very general and broad advantages of genetic diversity (in non-neuteral genes!) to trust or to the amount of scientific publications. (And why to choose exactly these two variables why not some other, related to cooperation and creativity?). For me, this is all far too unspecific and speculative.

    In sum, I do not believe in the story of the paper. But nevertheless I think, nevertheless it is not that bad as postulated by the anthropologist.
    Actually, the empirical strategy and the arguments in the paper are much more complex, differentiated and careful than in most papers written by economists.
    They control for almost every imagineable factor, using different variants, measures and adjustments of their variables and data. They worked on the paper for a quite long time and I think they really believe their story. Oded Galor always has unconventional and deep ideas and I think he is one of the most creative and methodologically best trained economist alive. Even more, he is one of the few economists that is willing to do interdisciplinary work – something strongly necessary to solve some the problems mainstream economics have today.
    To do interdisciplinary research is always a difficult undertaking. Blaming Ashraf and Galor for having the courage of trying it is neither helpful nor fair in my opinion.

    • Evans says

      C.W., I attended a recent lecture by Oded Galor on the subject. I raised a similar puzzle: how diversity in “junk DNA” can affect innovations and trust? The answer was convincing. Here is the relevant paragraph in the paper about it:

      It is relevant to note that the expected heterozygosity measure in the sample of 53
      HGDP-CEPH ethnic groups is based on microsattelites, i.e., DNA loci in nonprotein-
      coding regions of the human genome that do not directly result in phenotypic
      expression. Therefore, this measure of observed genetic diversity has the advantage
      of not being confounded by the forces of natural selection that may have operated
      on these populations since their prehistoric exodus from Africa. Importantly, however,
      the effects associated with heterozygosity in microsattelites capture the effects
      of diversity in phenotypically-expressed genomic material since the serial-founder
      effect, associated with the “out of Africa” migration process, is indeed reflected in
      other dimensions of within-group diversity, including diversity in various craniometric
      traits (Manica et al. Nature 2007).

  4. C.W. says

    @ Jason:

    Okay, do this. I don’t mind. Nice that like the comment!

    @ Evans:

    So, the argument is, that the diversity captured in the data, althought it reflects “Junk DNA” is associated with phenotypically-expressed diversity?
    Okay, this is a first step to convince me, but only a first.
    Why phenotypical diversity per se should matter? Does it matter for trust whether people are smaller or larger or do have lighter or darker hair? My point is, someone should come up with evidence that diversity in phenotypical attributes relevant for trust (like according to some experimental economic papers e.g. mimic/ facial expressions) and associated with particular genes matters for trust and therefore eventually for economic development. But since such type of genetic data is probably not available I guess (maybe out of good reasons!).

    Additionally, I want to add something to the critique of the anthropolgist concerning the use of McEvedy and Jones’ population data. In a paper published in AER in 2011 (Dynamics and Stagnation in the Malthusian Epoch) Ashraf and Galor do cross checking the validity of that estimates (see footnote 14 on p. 2011) and found that

    “…a recent assessment (see, e.g., www.
    census.gov/ipc/www/worldhis.html) conducted by the US Census Bureau finds that their aggregate estimates indeed compare favorably with those obtained from other studies. Moreover, the regional estimates of McEvedy and Jones are also very similar to those presented in the more recent study by Massimo Livi-Bacci (2001).” (Ashraf and Galor 2011, p. 2011) Again, I don’t see a problem with this data.

    Furthermore, the critique that they do not use the individual level trust data, albeit this is available is a little bit stupid, because Ashraf and Galor conduct a cross-country study and therefore cannot use individual level data – furthemore the aggregated values are available form the WVS website and aggregated in a reasonable way I suppose. My impression is, that the critique of the data is added by the authors simply to fill pages.

    • Evans says

      @ C.W.

      I found this response very convincing also.

      As to your follow up question, I suppose the reply of Asharf and Galor would be that even a thousand mile journey starts with one step. They find a negative reduced formed relationship between genetic diversity and trust and a positive one between genetic diversity and innovations. Future research will show it based on micro evidence. I think it is evident from our daily life. Isn’t xenophobia based on innate biased against dissimilar people? Isn’t it the best evidence that people that differ in their phenotypic characteristics (skin color, facial expressions, etc) tend to be less trustful for one another.

      The paper has a constructive policy implication. Education for tolerance and respect fro one another will permit society to enjoy the benefits of diversity.

  5. C.W. says

    @ Evans

    ” Isn’t xenophobia based on innate biased against dissimilar people? Isn’t it the best evidence that people that differ in their phenotypic characteristics (skin color, facial expressions, etc) tend to be less trustful for one another.”

    Of course, thats completely right.
    But , I don’t understand, is the genetic variation is skin color etc. contained in their data or not? If not, then I doubt that “some” genetic variation (in whatever) might be related to trust or to innovation. The anthropologists are right in claiming that they do almost completely neglect the large literature on trust (empirical and experimental) and on nature of human cooperation (coming from economists, but also from sociologist, evolutionary anthropolgists etc.). They also say not much about technological innovations. Why not using the index of technological adaption created by themselves in their 2011 paper or the similiar one constructed by Comin et al. (2010) for 0 , 1000 and 1500 AD (exactely the years considered by them). Why not doing something with the large literature on the historical persistence of technology? Why to use this “scientific publications” measure. I think at least to use this variable was not a very good choice.

    “…even a thousand mile journey starts with one step.”#

    You are completely right. That’s why I think, the paper isn’t that convincing until now, (as the things of Spolaore etc. are also not that convincing. Also they argue in a more specific way what is an advantage of these studies), but the idea behind and the potential of this way of research (and more broader of “unified growth theory” and very long-run development research) is very high. Because of this I think one should not be too hard with Ashraf and Galor. They do – in the light of the innovativness of the argument and method- a pretty good job! But again, in my opinion, the genetical economic literature has to solve some fundamental problems before I wiil believe in the results of that papers.

Comments, thoughts, suggestions?