Garett Jones has written an impressive new book on the implications of culture for international migration. In The Culture Transplant (subtitled: How Migrants Make The Economies They Move To A Lot Like The Ones They Left), Jones makes the following arguments:

1. Some cultures are better than others at certain important tasks, including good governance and wealth creation.

2. These cultural attributes relate to deep historical patterns, including the level of development achieved by that society’s ancestors thousands of years in the past.

3. These cultural attributes are relatively persistent and remain intact (to some extent) even several generations after individuals migrate to a new country.

Jones argues that these facts have important implications for immigration policy. In particular, he suggests that migration from cultures that are less successful will tend to degrade the receiving country. He worries about a decline in the quality of governance, less ability to innovate, and even (in extreme cases) more civil strife. I would expect this book to be of particular interest to conservative opponents of immigration.

Reviewers such as Bryan Caplan and Alex Nowrasteh have argued Jones overstates the case against high levels of immigration. I also believe that Jones somewhat overstates his case (which nonetheless may have some merit), and will suggest some additional reasons why.   

[Update:  I recently became aware of a second post by Alex Nowrasteh, done before mine.  He anticipates all my key points, and has a much more thorough discussion of the evidence.]

Jones writes in an engaging style and understands how to keep readers interested. Instead of opening with some dry theory, he begins with a few short chapters that provide extended anecdotes aimed at illustrating his basic point. Thus in the preface, Jones argues that poor countries such as Egypt, Paraguay, and Indonesia would benefit from receiving lots of immigrants from China. By beginning with this example, Jones is signaling that he’s not reflexively anti-immigration; rather he’s specifically worried about immigration from less successful cultures. China has a long history of achievement in three areas that Jones suggests are highly significant: state capacity (S), agriculture (A), and technology (T). He develops an index called SAT, which aggregates these metrics and assigns an SAT score for each country (not to be confused with the college entrance exam.)

In a later chapter, Jones shows that many Southeast Asian nations have benefited from Chinese immigration. I think he’s right, but I also believe he underestimates the problem with using China as an example of the importance of culture. If Chinese culture is so superior at wealth creation and good governance, then why is China itself relatively poor? And why has China been relatively poorly governed over almost all of the past 150 years? There are reasonable responses one can offer, and again I think he’s right about Chinese immigration to Southeast Asia. Nonetheless, it’s odd to start your book with such a poor example, a case that requires adding some “epicycles” to the model in order to explain awkward facts. Yes, China’s been held back by communism, but Jones’s argument is that some immigrants groups are better because they have cultures associated with good governance. So why has China been poorly governed for most of its recent history?

This is just a short opening chapter, and certainly doesn’t discredit his model, which relies on a wide range of empirical studies. But chapter one (“The Assimilation Myth”) also opens with an extended anecdote, which is even less favorable to Jones’s model. Jones explains how Argentina was one of the world’s richest nations back in 1913, but after a century of bad governments has fallen back to a middle-income level. He attributes their relative decline to a massive wave of immigration from Italy (and to a lesser extent Spain) in the early 1900s. (Although Spanish speaking, in an ethnic sense Argentina is dominated by Italians.)

[BTW, Razib Khan says that when people say something to the effect that “the model minority view of Asians is a myth,” it’s a pretty good indication that’s it’s at least partly true. That’s how I feel about “the assimilation myth.”]

The Argentine example has the opposite problem of the China example. Italy and Spain are fully developed Western European nations, with per capita GDPs that are nearly twice as high as Argentina (in PPP terms, there’s an even greater gap in dollar terms). So if these were low quality migrants, why do they produce such bad results in Argentina where they are only a portion of the population, and good results back in their European homelands, where they represent almost the entire population?

Italy is often cited as a case study for the cultural issues that Jones focuses on. Southern Italy has a relatively low trust culture with high levels of corruption and lots of low productivity family firms. Northern Italy has a much higher trust culture, with less corruption and many successful wealth-creating companies. So did Argentina receive its immigrants from the less successful part of Italy?  

Actually, only about half of Italian immigrants to Argentina came from southern Italy. In contrast, the overwhelming majority of Italian immigrants to the US came from southern Italy. And yet America’s Italian immigrants successfully assimilated into our middle class, while Argentina’s supposedly superior mix of Italian immigrants did poorly. Why? Jones mentions something about anarchist agitators arriving from Europe. But that sort of reliance on the effect of a few individuals with problematic political views is at odds with the sort of cultural determinism that underlies his model. (Here it might be helpful to recall the longstanding debate between the “great man” theory of history and the deep cultural forces theory. Jones is clearly in the latter camp.) 

After completing chapter one, we’ve encountered two important case studies: China and Italy. With China, we have immigrants leaving a dysfunctional society and doing well elsewhere, and with Italy we have immigrants leaving a highly successful society and doing poorly in Argentina. Again, this doesn’t mean Jones’s theory is wrong (I think he’s partly correct), but it’s a bit concerning that the two anecdotes he cherry picks to illustrate his model are such a poor fit for what he will subsequently try to show. He could have opened with any number of case studies, and indeed elsewhere he mentions better examples, such as the fact that Norwegians do well in Norway and also in the US.  

In chapter 5, Jones pushes back against the mantra that “diversity is our strength”.  He worries that cultural diversity can lead to civil strife and a deterioration in the quality of governance, which will eventually make a country poorer.  While this may be correct, it’s difficult to explain why highly diverse America is much richer than any other country with a population of more than 10 million.  (Our per capita GDP (PPP) is more than $5000 above second place Netherlands.)  If cultural diversity is a strong negative, how can the US be much richer than any other non-small country?  Why is America more than 50% richer than Japan? 

Some might dismiss the US case as a mere anecdote; what matters are the correlations that show up in statistical regressions involving many countries.  But the US is a fairly important case, and I suspect that most readers of Jones book will be Americans.  How do we know that the ability to assimilate immigrants is a stable parameter?  Casual empiricism suggests exactly the opposite.  East Asian immigrants seem to be assimilating relatively successfully into the US, while (as Jones points out) Chinese immigrants in Southeast Asia often maintain quite separate communities.  Muslim immigrants to America have done quite well, even as Muslim immigrants to France have done relatively poorly. 

Of course you can keep adding epicycles to the theory, such as looking at the specific type of Muslim immigrant to each country. But the nature of the receiving country is also important. In a dynamic market economy with a relatively small welfare state for non-workers, the level of employment is likely to be higher than in a more statist economy where unemployment is high and welfare benefits are strong. Employment is important, as immigrants who work with locals are more likely to adopt the local culture. I suspect that unemployed Muslim immigrants stuck in housing projects outside Paris are less likely to assimilate than fully employed Muslim immigrants in America.

On page 81, Jones does acknowledge that ethnic conflicts occasionally fade over time. I wish he had spent more effort thinking about why they tend to fade more in countries such as the US than in places like former Yugoslavia. I’d also have liked to see a discussion of Switzerland. Jones is famous for writing a book extolling the benefits of having “10% less democracy,” citing Singapore as an example. But Switzerland is an even more successful place than Singapore (if living standards are measured correctly), and it has at least 10% more democracy (and decentralization) than any other country. Might that be why the various language groups in Switzerland get along better than in former Yugoslavia? (Switzerland also has an extremely high proportion of immigrants.)

It is very difficult to predict the impact of immigration on a country’s politics. About 10 years ago, there were many predictions that immigration would make US politics more left wing. This was based on the fact that immigrants are more likely to vote for the Democrats. But this is a simplistic way of looking at the impact of ethnicity. Blacks tend to vote Democratic, but the higher the black population of a state, the more likely it is to be controlled by the Republicans. Immigration to the US seems to have energized the Republican Party, leading to the election of Trump in 2016.  (In my view, he would have been re-elected if he had been less . . . er . . . controversial.)

There is also evidence that ethnic diversity leads to a smaller welfare state and lower taxes, as the majority resists paying benefits to lower income minority groups. Some have argued that this explains why the welfare state in America is smaller than in Europe. It’s also been suggested that immigrants from places like Latin America will bring with them a preference for populist authoritarian leaders, the so-called “man on horseback.” But when America was finally presented with such a candidate in 2016 and 2020, it turns out that he was mostly supported by whites, and Hispanic voters tended to opt for the more liberal candidate. In big cities, whites are more likely to support ideas such as “defund the police,” while black voters shy away from these sorts of nutty ideas. In Democratic primaries, black voters tend to be more skeptical of candidates that identify as “socialist.” None of this means that Jones is wrong, but I suspect that the relationship between culture and politics is more complicated than he assumes.

In chapter six, Jones shows that most of the important innovations leading to higher living standards are produced by just a handful of major developed countries in Europe, North America and East Asia.  He worries that the quality of these countries may be watered down by mass immigration from less successful cultures, hobbling the primary engines of world innovation.  Oddly, he repeatedly suggests that being large makes a country more innovative, even though his own data doesn’t really support that claim.  For instance, while his data suggests that big countries produce the most Nobel Prize winners, it also shows that small countries tend to lead in Nobel Prize winners per capita.  Thus I don’t understand this claim (p. 118):

So Denmark, with a population of 5 million, probably uses many more ideas from Germany (population 83 million) than the other way around.  The small, in this way exploit the large—another reminder that every nation relies on the inventions created in just a few nations.

The term “exploit” seems misleading; he’s confusing total innovation with innovation per capita.  If Jones’s claim were correct, then if Germany were to divide up into a bunch of independent nations with the populations equal to places like Switzerland, Austria, Belgium, Denmark, Luxembourg and Netherlands, then global innovation would suffer—because bigger is better.  But in aggregate these six small nations are just about as successful in producing Nobel Prize winners (99) as is Germany (with 113), despite having a far smaller total population.  If anything, Germany is less innovative than its neighbors.  (Adding Sweden would boost the total of Nobel Prizes in this group of small nations to well above Germany, despite still having a lower total population.) 

In any case, the bigger is better argument is not necessary for Jones to make his point.  It’s enough to point out that at a global level innovation is concentrated in a few areas.  Even within the US, innovation is concentrated in places such as Silicon Valley, Boston and Hollywood.  The question is whether more immigration will hurt innovation, will “kill the goose that lays the golden eggs”.  Jones understands that innovations in places such as Silicon Valley are often produced by immigrants.  He’s making a different point.  He worries that mass immigration from unsuccessful cultures will degrade our political system, leading to worse economic policies and thereby reducing innovation in the long run.  But the book doesn’t present any examples of that phenomenon occurring, apart from the highly suspect example of Argentina.

I am not saying that Jones is completely wrong; indeed I suspect that his hypothesis is partly correct.  Culture does play an important role in the wealth of nations.  Cultural traits tend to persist over time.  But at the margin, I don’t see this concern as having important implications for US immigration policy, for several reasons:

1.  Immigrants to the US tend to assimilate better than immigrants to many other nations.

2. Immigrants to the US from even highly dysfunctional places such as South Asia and West Africa tend to do relatively well in the US.

Thus I see no evidence in Jones’s book contradicting the view that the US would benefit from any politically feasible boost in immigration.  So what are the policy implications?

1. Jones successfully raises some doubts about a policy of completely open borders, particularly for a small nation.  It’s hard for me to imagine the impact of Switzerland removing all border controls and allowing unlimited immigration from the world’s poorest nations.  And even for larger countries such as the US, that policy would have to be accompanied (at a minimum) by a removal of welfare benefits for new arrivals.  And even then the wave of immigration would probably be too much for the voters to accept, at least until world incomes become somewhat more equal. 

But not completely equal.  Contrary to popular imagination, not all income differences lead to large waves of migration.  Lots of people still live in places such as Bulgaria and Romania, despite free migration within the EU and income levels in the Balkans that are a small fraction of incomes in northwestern Europe.  But the world’s poorest countries are far poorer than even Bulgaria, and have vastly larger populations.  So completely open borders would be a very hard sell to the rich world’s voters.  

The second policy implication is that skill based immigration policies that you see in places like Canada and Australia may have more positive long run cultural effects than policy regimes that don’t favor high skilled immigrants.  To be clear, I don’t see any problem in the current mix of skills in US immigrants; they seem to be doing fine in most cases.  But there’s at least a respectable argument for shifting the US immigrant mix a bit further toward the highly skilled groups.  (Unlike Jones, I’m not convinced that the social science research on culture is strong enough to distinguish between relatively high skilled people from successful places like Norway and similar people from failed states like India and Nigeria, so I’d take them all.)

I’m surprised that Jones doesn’t spend more time discussing the advantages of cultural diversity.  The US entertainment industry (broadly defined to include film, comedy, music and sports) dominates the global scene.  Why is this?  I can’t help noticing that various minority groups play an important role in these industries (notably Jews and African-Americans.)  When people discuss the disproportionate share of Indian immigrants among Silicon Valley CEOs, they often point to cultural factors such as a familiarity with the English language (relative to East Asian immigrants.) It seems plausible that America’s diversity helps its economy by allowing various ethnic groups to engage in areas where each has a comparative advantage.  

In contrast, Japan did extremely well during the postwar decades when they focused on high quality manufacturing of consumer goods such as cars and TVs, but after the 1990s their monoculture proved unable to smoothly adapt to the post-industrial economy that relies heavily on creating new ideas that break with tradition.  An economy with cultural diversity might be less brittle, better able to adapt to a wide variety of economic conditions.

It seems to me that Jones’s book has implications that challenge some long held views on both the right and the left.  His research suggests that immigrants from less successful places are better off assimilating into American culture.  Yet the “identity politics” of the left increasingly opposes the traditional goal of making America a melting pot, and instead encourages groups to hold onto their ethnic identity.  Would Italian-Americans be better off today if they had held firmly to the cultural traditions of southern Italy?  Jones’s research does suggest that Italian Americans have not fully assimilated, but it’s certainly true that compared to when I was young one hears far fewer reports of the influence of the Italian mafia.  (In contrast, the mafia remains very active in southern Italy.)

If Jones is correct that cultures evolve extremely slowly over time, then conservatives may need to rethink their claim that the legacy of slavery does not provide an “excuse” for current problems in the African American community.  Recent West African immigrants that are doing well in America did not experience the brutal suppression of traditional family structures that occurred under American forms of slavery.  Conservatives cannot have it both ways, claiming that cultures are almost impossible to change in a period of 100 years, while also suggesting that America’s blacks should have simply rebuilt the cultural structures that were destroyed by slavery.

Bryan Caplan argues that even if Jones is correct, the actual implications of his book are that America should provide open borders with a fairly large range of countries, comprising a few billion people.  That would be a radical move toward significantly more open borders.  Caplan might well be correct, but it’s not the impression the typical conservative reader will take from the book.  I wonder how Jones responds to Caplan’s claim.

Alex Nowrasteh raises serious questions about the quality of the social science research that Jones relies on.  In a sense, I’ve implicitly raised some questions with my observations about America’s extraordinary success. We have a fairly substantial share of our population from places with somewhat dysfunctional cultures, such as Africa, Latin America, South Asia, southern Italy, Cambodia, Laos, and the Philippines.  And yet we are far richer than places with supposedly (culturally) superior populations, such as Japan and Germany.  Why?  Size alone doesn’t seem to provide the answer, as the smaller countries of northwestern Europe are roughly as rich as Germany, and some of the smaller economies in East Asia are roughly as rich as Japan (in PPP terms).  If Denmark, the Netherlands, Switzerland and Austria can all be richer than Germany, then why can’t Canada be richer than the US?  Perhaps our ethnic diversity is not the explanation, but it certainly doesn’t seem to have greatly held us back.  

A reader of Jones’s book might assume that many Americans are leaving dysfunctional states such as Texas for the greener pastures of West Virginia.  After all, Texas is only about 40% non-Hispanic white, and his model suggests that having a large share of people from places like Africa and Latin American will lead to bad governance.  In contrast, West Virginia is mostly white, with relatively few immigrants.  So why is it doing so much worse than Texas?

I would encourage people to check their biases.  I grew up as a white person in a heavily white area, and tended to view white culture as “normal”.  I wonder how many whites realize that Asian Americans often view whites as a violent gun-toting race?  Many whites are aware of the phenomenon of “white flight”, the tendency to move out of school systems with large black and Hispanic populations.  How many whites are aware of the existence of white flight from school systems where Asian students outperform whites? It’s human nature to view our own group as normal, having just the right amount of murder, just the right number of slacker students and meth addicts.  I don’t see evidence of that sort of bias in Garett Jones, but I worry it exists in the larger anti-immigrant community. Yes, immigrants sometimes bring problems.  But they also bring in a fresh set of skill and attitudes, which enrich the complex American mosaic.  

PS.  Don’t be put off by my objections to specific points made by Jones.  It’s an excellent book and well worth reading.  There are no easy answers in this area.