Bryan Caplan  

Ancestry and Long-Run Growth Reading Club: Spolaore and Wacziarg

Foglia on Piketty and inequali... Beckworth and Ponnuru on 2008...
Welcome to the third installment of the EconLog Reading Club on Ancestry and Long-Run Growth. This week's paper: Spolaore, Enrico, and Roman Wacziarg. "How Deep Are the Roots of Economic Development?" Journal of Economic Literature 51 (2): 325-369.  The authors' data is here.


Spolaore and Wacziarg survey a broad literature on long-run growth, paying special attention to measures of ancestry.  They begin by surveying the evidence on the strong long-run effects of geography: absolute latitude and islands are good for prosperity, tropics and being landlocked are bad.  They then cover and critique Acemoglu, Robinson, and Johnson's famous "reversal of fortune" paper.  According to SW, ARJ's results are fragile:
[W]hen Europe is included in the sample, any evidence for reversals of fortune disappears: the coefficient on 1500 population density is essentially zero for the broadest sample that includes the whole world (column 1). For countries that were not former European colonies, there is strong evidence of persistence, with a positive significant coefficient on 1500 density. The evidence of persistence is even stronger when looking at countries that are populated mostly by their indigenous populations (the evidence is yet stronger when defining "indigenous" countries more strictly, for instance requiring that more than 90 percent of the population be descended from those who inhabited the country in 1500).
This in turn leads SW into the work this Read Club has already covered.  But rather than simply cite them, SW tweak their regressions and report the results.  Log per capita income in 2005 is the dependent variable.

First, agricultural measures:


Second, state history measures:


Both sets of regressions reinforce the lessons of Putterman and Weil: ancestry and geography are both big deals.  If I'm correctly applying SW, though, they find a smaller effect of ancestry and a larger effect of geography, presumably because they add the "Island" variable.  An extra 10,000 years of ancestral agriculture boosts expected per-capita income by 269%.  But moving 25 degrees from the equator has the same effect.  And moving from a landlocked country to an island at the same latitude boosts per-capita income by 343%! 

For ancestral state history, similarly, going from the min to the max raises per-capita income by 337%.  But moving 26 degrees from the equator does the same, and moving from a landlocked country to an island at the same latitude boosts per-capita income by 288%.

SW then move into their own work on genetic distance.  What's genetic distance?
Genetic distance is a summary measure of differences in allele frequencies between populations across a range of neutral genes (chromosomal loci). The measure we used, FST genetic distance, captures the length of time since two populations became separated from each other. When two populations split apart, random genetic mutations result in genetic differentiation over time. The longer the separation time, the greater the genetic distance computed from a set of neutral genes. Therefore, genetic distance captures the time since two populations have shared common ancestors (the time since they were parts of the same population), and can be viewed as a summary measure of relatedness between populations.
Genetic distance from the United States turns out to be another highly-predictive ancestry measure - though, as usual, geography also matters enormously.

SW close the paper with a lengthy discussion of where the results come from and what they mean.  Most notably, they ponder the distinction between "direct" and "barrier" effects of long-run forces:
One possibility is that intergenerationally transmitted traits have direct effects on productivity and economic performance. A slow-changing cultural trait developed in early history could be conducive directly to high incomes in modern times if it is transmitted from parents to kids within populations, either behaviorally or through complex symbolic systems (e.g., by religious teaching). For example, a direct effect stemming from cultural transmission would be Weber's (1905) argument that the Protestant ethic was a causal factor in industrialization (a recent critical reassessment of this hypothesis has been provided by Becker and Woessmann 2009).

As we discussed in the previous section, another possibility is that human traits act to hinder development through a barrier effect. In this case, it is not the trait itself that directly affects economic performance. Rather, it is differences in inherited characteristics across populations that create barriers to the flow of technological and institutional innovations, ideas, etc., and, consequently, hurt development. Historically rooted differences may generate barriers--e.g., via cultural, racial, and ethnic bias, discrimination, mistrust, and miscommunication--hindering interactions between populations that could result in a quicker diffusion of productivity enhancing innovations across populations...
SW close by trying to prevent misinterpretation of the research.  Yes,
Taking the recent literature seriously implies acknowledging the limits faced by policymakers in significantly altering the wealth of nations when history casts a very long shadow. A realistic understanding of the role of historical factors is essential for policy assessment. One could obtain misleading conclusions about the effects of specific policies and institutions when not taking into account the role of long-term variables.
But the fraction of variance explained by long-run forces is way below 1, and many countries have broken out of bad long-run trajectories.

Critical Comments

1. Overall, I have less to say about SW because most of my earlier critical comments transfer over.  Besides introducing the economics profession to a lot of interesting research, SW's main value-added is providing outside verification that ancestry regressions are robust.  All the main results we've reviewed hold up: ancestry matters a lot, even controlling for geography, but geography is also greatly important, controlling for ancestry. 

2. SW's caveats are understated. All their regressions - and as far as I can tell, all the research they discuss - try to explain why some countries are richer than others.  None of them even whisper that some countries are doomed to absolute poverty.  The amazing economic history of the last two centuries shows the opposite.  While relative performance is fairly stable, absolute performance has skyrocketed.  Remember last week's graph:


The same holds if we put any of the key ancestry measures on the x-axis: In any given year, countries with more successful ancestors do better.  Over time, however, the whole success curve is rapidly shifting up.

To be fair, SW are writing in an intellectual environment where the basic facts of exploding global prosperity are well-known.  But non-economists who read SW are dangerously likely to embrace a misguided fatalism.

3. Once again: If we take all of SW's results seriously - ancestry and geography - the case for free migration gets stronger, not weaker.  Global economic growth will end absolute poverty eventually, but all this research confirms that moving the world's poor to rich countries is an amazing short-cut. 

COMMENTS (4 to date)
Martin Ljunge writes:

Thanks for the review Bryan!

My main quandary or confusion with the SW paper is their distinction between "direct" and "barrier" effects. Is it possible to empirically distinguish between these? More specifically, can one convincingly argue for a "direct" effect on productivity of some trait while excluding that it works through a "barrier"?

For example, Algan and Cahuc (2010, AER) finds that inherited trust predicts development. This could work both directly (trusting individuals realize more productive opportunities) and through barriers (trusting individuals adopt new tech faster). But what kind of data would distinguish the direct effect from the barrier effect? Any evidence I can think of that point to a direct effect could have a barrier interpretation. It is probably just my limited imagination but the distinction seems hard to make empirically. Please add your comments to this distinction if you have a chance.


Enrico Spolaore and Romain Wacziarg writes:

Thank you, Bryan, for your thoughtful overview of our paper. We agree with you that the results of this literature should not be interpreted fatalistically. As we write in the conclusions of our paper, the fact that current development is a function of long-term factors does not imply that we are hopeless. On the contrary, long-term history is not a deterministic straightjacket, and the historical evidence is consistent with some optimism about our ability to overcome long-term constraints. That’s for several reasons. First, historical variables do not explain all of the variation in income per capita. And while there is significant persistence in the use of advanced technologies over time and space, there have also been significant shifts in the technological frontier, with populations at the periphery becoming major innovators, and former frontier societies falling behind – a fact that you correctly notice in a previous post. Secondly, the diffusion of knowledge takes place not only vertically – across generations within a population - but also horizontally, across different populations. Through horizontal transmission and cultural change, major changes can take place relatively quickly, sidestepping historical constraints. We like a famous quote attributed to Senator Daniel Patrick Moynihan: “The central conservative truth is that it is culture, not politics, that determines the success of a society. The central liberal truth is that politics can change a culture and save it from itself.”

We believe that we can be especially optimistic if we interpret the effects of history and ancestry in terms of barriers to the diffusion of development and innovations. Barriers can be overcome and have indeed been overcome historically. In fact, in our own work we find that the effect of relative genetic distance on income differences peaked in the second half of the nineteenth century, and has been declining in the second half of the twentieth century. This suggests that the impact of vertically-transmitted traits has changed in the era of globalization, through greater exchanges of human capital, ideas, blueprints, and greater convergence of norms and values, facilitating the horizontal diffusion of innovations. In your blog posts you often point out the economic benefits of free movements of people across borders. From the perspective of our research, such movements could be especially beneficial if they reduce barriers to the exchange of ideas and productivity-enhancing innovations across different cultures and societies.

Richard writes:
One possibility is that intergenerationally transmitted traits have direct effects on productivity and economic performance. A slow-changing cultural trait developed in early history could be conducive directly to high incomes in modern times if it is transmitted from parents to kids within populations, either behaviorally or through complex symbolic systems (e.g., by religious teaching).

Because we all know that the only ways that children acquire traits are from parents or society.

Chuck writes:

"Global economic growth will end absolute poverty eventually, but all this research confirms that moving the world's poor to rich countries is an amazing short-cut."

Research by Heiner Rindermann has shown that the intelligence of countries' smart fraction can explain a large chunk of outcome differences. This suggests a simple, more egalitarian, more efficient alternative to the dilemma: encourage a large chunk of the rich countries' cognitive elite to emigrate to poor countries -- an embodied human capital transference. Surely, the rather apt open borders crowd wouldn't mind volunteering for the sake of humanity! If not, I'm sure that they could be encouraged to go --. I say more egalitarian because this smart fraction transfer would have an impact on the whole of the poor countries' population, not just the portion which will manage to migrate and also because it respects the right of self-determination of the rich nations. Not everyone can live in island countries at high absolute latitudes. There simply isn't enough room. (Though the idea of emptying out e.g., Africa and leaving it as a massive wildlife preserve has its appeal.) So it would be more efficient to find a way to improve the outcomes of the supposedly geographically impaired countries. And what I suggest is such a way.

Comments for this entry have been closed
Return to top