Welcome to the first installment of the EconLog Reading Club on Ancestry and Long-Run Growth. This week’s paper: Putterman, Louis, and David Weil. 2010. “Post-1500 Population Flows and the Long-Run Determinants of Economic Growth and Inequality.” Quarterly Journal of Economics 125(4): 1627-1682.  The authors’ data is here.


Putterman and Weil start by noting that – at least by some measures – economic success is persistent over the centuries.  Countries that were advanced in the distant past tend to be richer today.  But should we think of countries as locations or peoples?  Economists routinely do the former, but maybe they shouldn’t.

[T]he further back into the past one looks, the more the economic history of a given place tends to diverge from the economic history of the people who currently live there. For example, the territory that is now the United States was inhabited in 1500 largely by hunting, fishing, and horticultural communities with pre-iron technology, organized into relatively small, pre-state political units. In contrast, a large fraction of the current U.S. population is descended from people who in 1500 lived in settled agricultural societies with advanced metallurgy, organized into large states. The example of the United States also makes it clear that, because of migration, the long-historical background of the people living in a given country can be quite heterogeneous.

To surmount this problem, P&W laboriously construct a country-by-country matrix of ancestry:

We construct a matrix detailing the year-1500 origins of the current population of almost every country in the world. In addition to the quantity and timing of migration, the matrix also reflects differential population growth rates among native and immigrant population groups. The matrix can be used as a tool to adjust historical data to reflect the status in the year 1500 of the ancestors of a country’s current population. That is, we can convert any measure applicable to countries into a measure applicable to the ancestors of the people who now live in each country.

How could one even begin to construct such a matrix?  Whenever possible, P&W use actual genetic data, then supplements genetics with history.  What does their matrix look like?

The matrix has 165 rows, each for a present-day country, and 172 columns (the same 165 countries plus seven other source countries with current populations of less than one half million). Its entries are the proportion of long-term residents’ ancestors estimated to have lived in each source country in 1500. Each row sums to one. To give an example, the row for Malaysia has five nonzero entries, corresponding to the five source countries for the current Malaysian population: Malaysia (0.60), China (0.26), India (0.075), Indonesia (0.04) and the Philippines (0.025).

The resulting matrix exposes two basic facts. 

1. Long-distance migration is rare.  Most modern countries are almost entirely populated by the descendants of earlier inhabitants of the region. 

2. Long-distance migration is bimodal.  When countries aren’t almost entirely populated by the descendants of earlier inhabitants of the region, those earlier inhabitants usually have almost no descendants left in the country.

Check out the Distribution of Countries by Proportion of Ancestors from Own or Immediate Neighboring Country:


Here’s the Distribution of World Population by Proportion of Ancestors from Own or Immediate Neighboring Country:


P&W now revisit earlier research on long-run growth, using their matrix to transform the key variables.  Instead of looking at the long-run effects of places’ traits, they look at the long-run effect of tribes’ traits.  They focus on two measures: state history and years of agriculture.  State history measures how long a country “had a supratribal government, the geographic scope of that government, and whether that government was indigenous or by an outside power.”  Following previous work, they massage this measure: “The version used by us, as in Chanda and Putterman (2005, 2007), considers state history for the fifteen centuries to 1500, and discounts the past, reducing the weight on each half century before 1451-1500 by an additional 5%.”  Years of agriculture, in contrast, is not massaged.  It’s simply the “the number of millennia since a country transitioned from hunting and gathering to agriculture.”

Since migration history is bimodal, adjusting these measures for migration has a bimodal effect.  Most countries are near the 45-degree line, but a few radically diverge.  Here’s state history by location versus people:


Here’s agricultural history by location versus people:


Now for the punchline: Migration-adjusted measures are much more predictive of modern GDP than raw measures.  “Not surprisingly, given previous work, the tests suggest significant predictive power for the unadjusted variables. However, for both measures of early development, adjusting for migration produces a very large increase in explanatory power. In the case of statehist, R2 goes from .06 to .22, whereas in the case of agyears it goes from .08 to .24. The coefficients on the measures of early development are also much larger using the adjusted than the unadjusted values.”  Regression tables:


P&W try an array of robustness tests.  The most notable challenge they address, however, is “What about geography?”  Earlier researchers have found strong effects of latitude and being landlocked.  Correcting for these factors, the effects of migration-adjusted history crash, but remain absolutely large.


After further checks, P&W switch gears to analyze how ancestry matters for current inequality.  Punchline:

[T]he heterogeneity of a country’s population in terms of the early development of its ancestors as of 1500 is strongly correlated with income inequality. We also show that heterogeneity with respect to country of ancestry or with respect to the ancestral language does a better job than does current linguistic or ethnic heterogeneity in predicting income inequalities today.

Critical Comments

This is an awesome paper.  While I’m sure Putterman and Weil made countless judgment calls in constructing their migration matrix, I’d bet that if I re-did their work, my matrix would have at least a .8 correlation with theirs.  This is a classic “obvious once you think about it” paper, because rich non-Eurasian countries are now largely inhabited by Eurasians.  It is also a courageous paper.  As far as I know, P&W were not tarred and feathered for racism, but they sure could have been.  All political correctness aside, though, what is the reasonable way to interpret their results?

1. While Garett Jones sees ancestry research as very damaging to the case for free migration, Putterman and Weil acknowledge that most historical migration was anything but free.  Indeed, demographic change largely reflects conquest, genocide, and slavery.

Conquest, colonialism, migration, slavery, and epidemic disease reshaped the world that existed before the era of European expansion. Over the last 500 years, there have been dramatic movements of people, institutions, cultures, and languages among the world’s major regions.

In other words, civilized migration – where people voluntarily move to a new country to peacefully improve their lives – is an extreme historical rarity.  While this doesn’t prove that civilized migration has better long-run effects than conventionally brutal migration, it is plausible that dramatic long-run differences exist.  Most obviously, civilized migration is an effective way to sincerely culturally “convert” people – and their children.  Enslaving them, not so much. 

2. P&W also detect a surprising benefit of ancestral heterogeneity:

We find that, holding constant the average level of early development, heterogeneity in early development raises current income, a finding that might indicate spillovers of growth-promoting traits among national origin groups.

3. While it doesn’t affect their case, P&W use extremely low populations for the New World in 1500.  Is 1491 – and all the research upon which it builds – wrong?  According to Wikipedia’s summary of current research, P&W are off by a factor of three or more.

4. The time discounting of state history is worrisome.  Putterman and Weil use Chanda and Putterman’s 5% per half-century discount rate, but was this rate cherry-picked to make the results come out right?

5. On Twitter, Garett has been using the literature P&W jump-started to dissuade Western countries from admitting Syrian refugees.  But by Putterman and Weil’s agricultural measure, Syrians (and every else in the Fertile Crescent) are the most awesome people on the planet.  Here’s who the Syrians are, in case you’re curious:

The vast majority of Syria’s population is Arab, and is assumed to be indigenous to the country. There are also some Palestinian (2.8%) and Lebanese (0.5%) refugees living in the country (NE, WCD). Some Kurds (4%) have lived in the country for generations, but others (4%) came from Turkey in the early 20th century (LC, EV). The majority of Syria’s Armenians (0.2%) and Turks (0.3%) also came from Turkey during the first half of the 20th century as refugees (LC, EV,WCD). The Bedouin Arabs (7.4%) are believed to have emigrated from the interior of the Arabian peninsula between the 14th and 18th centuries; thus, the ancestors of about two-fifths of the Bedouins are treated as having lived in Saudi Arabia (3%), which constitutes the vast majority of the peninsula. There is also a small population of Azerbaijanis (0.7%) whose ancestors are assumed to have lived in present-day Azerbaijan (WCD).

Azerbaijan: 0.7%
Israel/Palestine: 2.8%
Lebanon: 0.5%
Saudi Arabia: 3%
Syria: 88.5%
Turkey: 4.5% (4% as ancestors of Kurds, 0.2% as ancestors of Armenians)

6. P&W are writing in a field where the effects of geography are well-established.  They aim to show that ancestry matters even controlling for geography – and they succeed.  But geographic variables remain highly potent.  Who cares, when you can’t change geography?  Well, you can’t change the geography of places, but you can change the geography of people.  How?  Migration! 

Look at column (6) of their Table IV above.  State history ranges from 0-1.  Moving from the minimum to the maximum state history score predicts a +1.24 change in log GDP – an increase of about 350%.  But you can get the same bonanza by moving 37 degrees away from the equator.  If you go from a landlocked country to a non-landlocked country, a 20 degree move suffices. 

Results for agriculture are similar.  The variable ranges from 0-10.5.  Moving from the minimum to the maximum agriculture score predicts a +1.61 change in log GDP – an increase of about 500%.  You can get the same bonanza by moving 40 degrees away from the equator – or 26 degrees if you journey from a landlocked country to a non-landlocked country.

Taking P&W’s results literally, then, this is a radically pro-immigration paper.  Let mankind move away from the tropics and toward the coasts, and the predicted net effect is cornucopian.  This would be true even if all migrants’ ancestors were complete savages in 1500 AD.  And if you look at P&W’s numbers, you’ll learn that even sub-Saharan Africans were well-above the minimum by 1500 AD, with state history scores around .2 and agriculture scores around 2.  The implied long-run payoff from relocating humanity from poor countries to rich countries is plausibly even higher than the “double global GDP” estimate Michael Clemens popularized in “Trillion-Dollar Bills on the Sidewalk.”

7. P&W seems to imply a NIMBY result for immigration.  Sure, migration enriches mankind, but doesn’t it reduce per capita GDP in receiving countries?  Taken literally, their full results imply the opposite, but P&W are skeptical:

The coefficients also have the unpalatable property that a country’s predicted income can sometimes be raised by replacing high statehist people with low statehist people, because the decline in the average level of statehist will be more than balanced by the increase in the standard deviation. For example, the coefficients just discussed imply that combining populations with statehist of 1 and 0, the optimal mix is 86% statehist = 1 and 14% statehist = 0. A country with such a mix would be 41% richer than a country with 100% of the population having a statehist of 1…

Although our regression result reflects the fact that population heterogeneity has not detracted from economic development in the first group of countries, it seems best not to infer from it that “catch up” by homogeneous Old World countries would be speeded up by infusions of low statehist populations into existing high statehist countries.

Econometrics aside, per-capita GDP is a dreadful measure of national welfare when migration is high.  As I often point out, immigration can enrich everyone in a country while reducing per-capita GDP.  In any case, the NIMBY inference depends on the size of the receiving country.  If Belgium doubles its production via migration, most of the extra production will spill over onto the rest of the world.  But if the EU doubles its production via migration, most of the extra production will probably be enjoyed by people in the EU.

8. Overall, this is an amazingly edifying paper.  If your main reaction is name-calling, you don’t belong in the world of ideas.  If the paper showed that my crusade for open borders was misguided, I would just have to live with that unwelcome conclusion.  But it shows nothing of the kind.  Immigration critics looking for intellectual foundations will have to keep looking – or ignore half the paper’s results.