Garett Jones’ new working paper on “Measuring the Sacrifice of Open Borders” has already received much attention. Before I respond to it, though, let me recount The Story So Far.
1. About a decade ago, researchers such as Michael Clemens started using standard trade models to estimate the economic effects of open borders. All such estimates were enormous, giving rise to slogans like “Open Borders: “The Efficient, Egalitarian, Libertarian, Utilitarian Way to Double World GDP.”
2. In 2015, my colleague Garett Jones published Hive Mind: How Your Nation’s IQ Matters So Much More Than Your Own, arguing that national IQ has a large causal effect on national outcomes. It’s a great book; here’s my review with further commentary.
3. Although Hive Mind makes many optimistic observations about immigration, readers routinely interpret the book as an attack on low-skilled immigration. Open borders wouldn’t enrich the world; instead, it would impoverish the world by breaking the countries that work.
4. Garett subsequently wrote a presentation sympathetic to the latter interpretation, entitled “Are the Global Benefits of Open Borders a Fallacy of Composition?” (though he still primarily relied on “Deep Roots” models rather than national IQ).
5. Caplan and Weinersmith’s new Open Borders: The Science and Ethics of Immigration used Hive Mind‘s preferred estimates to run a little thought experiment: What would happen if everyone on Earth moved to the U.S., bringing our national IQ down to the global average? The result: Global output still roughly doubles.  Here’s the key page, featuring me as a virtue-signaling knight debating a Jonesian alien:
6. Last week, Garett published a new working paper critiquing my thought experiment. He confirms all my math: In a fully Jonesian framework, open borders is still a fantastic opportunity for mankind. GDP per capita falls by about 50% in rich countries, but total production rises by 81%.
7. So what’s the rumpus? Garett argues that more than 100% of the gains will go to immigrants! So even though open borders nearly doubles the production of mankind, it reduces living standards of the current inhabitants of rich countries by a massive 40%. In short, we have a classic NIMBY (“Not In My Backyard”) picture; open borders is great for humanity, but nevertheless awful for us.
8. How is this possible? Drawing on earlier work, Garett insists that the personal payoff for IQ is modest. 1 IQ point raises earnings by about 1%. Since current U.S. IQ is about 11 points above the world average, the current citizens of rich countries will end up earning roughly 60% of what they now earn. In other words, Garett’s concern is that under open borders, income will be too equal for current residents of high-IQ countries to maintain their standard of living.
9. “Measuring the Sacrifice of Open Borders” doesn’t openly declare the sacrifice excessive, but if this isn’t a Straussian passage, what is?
But using Caplan’s base year estimate of U.S. income per capita of $58,000 per person, that would suggest the mean per-capita income of ex-ante Americans would fall from $58,000 to $33,000. It’s about 14% higher than the Open Borders America average of $29,000, but quite a decline from the world before Open Borders. Is a 43% drop in income a large sacrifice for the cause of Open Borders? I’ll leave that for each reader to answer personally.
So is Garett right?
1. To begin, reflect on how implausible Garett’s scenario is. In his story, the production of mankind virtually doubles, yet the average person in rich countries receives less than none of the gain. Why is this implausible? Because throughout history, large increases in production have invariably been broadly beneficial. The Industrial Revolution didn’t just benefit factory owners; it benefited everyone who consumed the mountains of new products. The internet didn’t just benefit programmers and tech entrepreneurs; it benefited everyone who surfs the web. Vaccines didn’t just benefit medical researchers; they benefited everyone who would have died without them. While progress always harms someone, the idea that global GDP could rise by 81% without trickling down to the current residents of the First World is bizarre.
2. Bizarreness aside, what should we think about Garett’s “favorite estimate” that 1 extra IQ point raises wages by a mere 1%? The natural question is: controlling for what? As Garett himself has emphasized in the past, you should beware of “Everest regressions.”  As the quip goes:
Controlling for barometric pressure, Mount Everest has the same altitude as the Dead Sea.
Thus, if IQ causes higher earnings both directly and by raising education, controlling for education masks much of the true effect of IQ.* When I simply regress log income on AFQT percentile with no controls, I get a coefficient of .011, which implies that being 11 IQ points (or 27 percentiles) above your national average raises earnings by .3 log-points. This is hardly the only way to run the estimates, but it’s reasonable, and implies much larger private benefits of IQ – and much smaller native losses – than Garett’s preferred figure. My underlying suspicion, to be frank, is that (unlike Garett himself) most economists are IQ-phobic and have therefore searched mightily for ways to downplay the personal payoff for IQ.
3. The deeper question, though, is: Should we expect the modest private benefit of IQ we see in current U.S. data to persist under open borders?
The right answer is, “Clearly not.”
Under open borders, the high-IQ share of the population shrinks. Due to this fall in relative supply of high-IQ workers, we should expect the market reward for IQ to rise. By how much? Consider South Africa. 80% of its citizens are black, 9% white, 9% mixed, and 2% Asian. White IQs far exceed black IQs. If Garett’s results for the private benefits of IQ were constant, we’d still expect tiny racial earnings gaps. Yet almost three decades after the end of apartheid, white earnings in South Africa far exceed black earnings; the average white makes about 500% more than the average black, and 250% more than the national average. If you do the math, a private payoff in this ballpark implies that open borders is no “sacrifice” for natives. Instead, like every other previous massive increase in human productivity, open borders is a widely-shared bonanza. Just as my new book says.
4. I fear that most pro-immigration readers will casually believe me, and most anti-immigration readers will casually believe Garett. Before you make up your mind on the Big Picture, though, you really should reflect on the underlying building blocks of our arguments. Would IQ have a big effect on personal economic success under open borders? Would there be high inequality under open borders? If you answered Yes to both questions, you should be on my side. If you answered No to both, you should be on Garett’s side. Despite mood affiliation, that’s the truth.
5. As I explain at length in Open Borders, there are strong reasons to believe that open borders would sharply raise the IQs of the children of immigrants without depressing the IQs of the children of natives. So even if the market payoff for IQ were unrelated to its supply, Garett’s estimated effects on natives are way too pessimistic, because national IQ will fall much less than naive demographic arithmetic implies.
6. As I told Garett during our 2016 debate on this topic, his case against open borders is still implicitly a case for radical liberalization of immigration. He has every reason to support open borders with China, Japan, Korea, Canada, Australia, New Zealand, and Europe.* His own models predict that this would be fantastic for natives and immigrants alike. Why Garett chooses to market himself as an immigration skeptic remains a mystery to me – and why so many immigration skeptics accept him as one of their own is a riddle, wrapped in a mystery, inside an enigma.
* And if Jones bases his case on “Deep Roots” rather than national IQ, he should also support open borders with India, Pakistan, Bangladesh, and the Middle East, all heirs of civilizations of great antiquity.
READER COMMENTS
Christophe Biocca
Nov 19 2019 at 10:32am
That asterisk doesn’t have a counterpart at the bottom of the page.
Josh
Nov 19 2019 at 10:40am
I feel like Jones’s model that high average IQ leads to increased wealth is off. The model I have in mind is that adding high IQ people to an area tends to increase the wealth of that area (because the high IQ people invent things that increase productivity, etc.). And adding high IQ people to an area also obviously raises average IQ in that area because of algebra.
So there is a strong correlation between average IQ and wealth as Jones notes and it’s in the direction of IQ->wealth. But it would be a mistake to then conclude that adding low IQ people brings down that wealth.
As an example, if everyone in your society has an IQ of 100, would it really get poorer if you admitted 1 person with an IQ of 140 and 9 with an IQ of 90 (average newcomer has IQ 95). Jones seems to predict the country will get poorer because the social return is determined by average IQ. I predict it will get richer because the high IQ people will raise the productivity of everyone (eg by inventing things).
I haven’t read Hive Mind so I don’t know if he refutes my model, but I find it implausible that adding low IQ people to an area drastically lowers the productivity of the people who are already there.
Floccina
Nov 19 2019 at 1:03pm
That is one of my thoughts. Is it high average IQ or having a lot of very high IQ people?
anonymous
Nov 19 2019 at 3:43pm
Compare India to China?
AMT
Nov 19 2019 at 11:48pm
I agree with Josh and was thinking the same thing.
M, your analysis is incorrect, because the benefit from a high number of high IQ people could be offset by a large number of low IQ, less productive people. I.e. the analysis shouldn’t focus on national per capita GDP, but GDP for each individual compared to the alternative (or the opportunity cost). The smart people invent things that benefit everyone. Having less productive/lower IQ people benefit highly productive people because there is more specialization and trade. I’m pretty sure we can agree that some bad institutions can suppress a nation, no matter how many high IQ people it has (e.g. communism).
But according to Garett’s logic, it would be best for all the smartest people to live in their own community, purely to keep that average IQ higher. He apparently believes they would have not only higher GPD per capita, but the geniuses in their own community, without any low IQ janitors, mechanics, or service sector in general, will be more productive than they were before, when they could hire people to cut their hair, fix their car, mow their lawn, cook their food, etc.!(obviously they could still benefit from trade in goods, at least) He is an economist, that has somehow forgotten the basics we have known since Adam Smith and David Ricardo.
Ok that might be taking Garett’s argument slightly further than even he would, but he must literally believe that these massive gains from specialization and trade must, somehow, be completely wiped out, and then some. I would like to know exactly how the mere existence of below average IQ people in society personally harms me far more than it benefits me. Welfare? Crime? Voting in Elizabeth Warren? Voting in Donald Trump? And none of Bryan’s suggested solutions to those concerns are viable?
Emily
Nov 19 2019 at 10:45am
I don’t casually agree with either of you all because if your thought experiment of “What would happen if everyone on Earth moved to the U.S” doesn’t involve the collapse of the United States government far before that can happen, we don’t have a lot to say to each other. This “who do the economic gains go to” is a quibble in the larger discussion of what happens to our institutions, and both of you all are assuming everything keeps pretty much just working.
Also, why is high income inequality that benefits workers like me supposed to be some kind of selling point? I actually don’t want to live in a de facto racialized caste system, even if I get to be in the favored caste, and I don’t want to raise my kids in it. If that’s NIMBY-ism, I’m cool with that, and I think a lot of other people are as well. Using South Africa as a positive example is going to be horrifying for both the right and the left.
Matthias Görgens
Nov 19 2019 at 11:05am
Inequality in Bryan Caplan’s suggested scenario would be high initially, but much lower than today. And the poorest would be richer.
anonymous
Nov 19 2019 at 3:43pm
Not within the U.S.
Daniel Hill
Nov 19 2019 at 11:14am
” I actually don’t want to live in a de facto racialized caste system”
Emily, you already do, the only difference is that those who get the short end of the stick are in other countries where you don’t have to think about them. “I don’t want to be surrounded by poor people, so let them be even poorer somewhere else” doesn’t seem a morally admirable position.
Emily
Nov 19 2019 at 11:32am
Perhaps Bryan could just lead with that, then. “It doesn’t really matter if you think that open borders would lead to life being dramatically worse for you, your family, and community — it’s immoral not to do it.” But since this is his attempt to persuade by making a different argument, I am sharing why I don’t find it persuasive in part because the values being appealed to are not my values.
robc
Nov 19 2019 at 12:08pm
I am pretty sure your preferred argument is also in the book.
Emily
Nov 19 2019 at 1:00pm
If it’s like the Foreign Policy article, I would say my argument is referred to, but that doesn’t mean it’s answered. For instance, Bryan tells us to worry less about terrorism:
That average is covering some pretty major variation, and at any rate, measuring terrorism in lives lost misses huge costs. How many acts of terrorism on the scale of 9/11 or even an order of magnitude less before we’re over as a liberal democracy? ‘Worrying about terrorism really is comparable to worrying about lightning strikes’ is not exactly a deep level of grappling with this.
AMT
Nov 19 2019 at 10:56pm
Where in the above post do you see Bryan saying “open borders would lead to life being dramatically worse for you, your family, and community”? I gathered…the exact opposite…
Alexander Turok
Nov 19 2019 at 5:45pm
I always think of South Africa when I think about open borders. Suppose that there were no Blacks in the country. Would South Africa’s whites benefit from allowing massive Black immigration from the rest of the continent? On the one hand, they would enjoy much cheaper services on account of being able to import cheaper labor from their neighbors. It’d be cheaper to build a new home there, but then, you realize you need a gate around your neighborhood, and you need to pay more to live and send your kid to school away from the immigrants.(Yes, you could save money by not being ‘racist,’ but most whites in nearly every region are.) Then you get into the issue of governance. There’s the keyhole solution of not letting the immigrants vote, but how long can that last? Of course Blacks are the natives there, and this hypothetical can’t map onto the actual conditions.
If you look at people voting with their feet, a lot more whites are leaving the country than entering. There is supposed to be a massive gain from highly skilled and less skilled workers being in close proximity. Comparative advantage and all that. But you don’t see much emigration of high-skilled people to low-income countries to take advantage of the gains of that proximity. What you do see tends to be retirees, middle-aged men, not those with families.
In the internal American context, levels of immigration don’t seem to have much impact on where Americans prefer to live, California sees very high levels of domestic outmigration consistent with fleeing diversity, while Texas sees very high levels of domestic inmigration consistent with seeking out diversity.
AMT
Nov 19 2019 at 11:14pm
If you believe Bryan, global inequality will go down, because the world’s poorest get much richer, and the richest get somewhat richer. If you believe Garett, global inequality will go way down, because the world’s poorest get much richer.
So, I’m not sure if you actually read any of his posts or his book, but it sounds like you would fully support open borders!
The world that restricts people’s movement and ability to take jobs from willing employers, resulting in far higher inequality, where the primary determinant of that inequality is nationality, sounds a lot more like a racialized caste system than one with more freedom…
Uh…so your values include “Screw the poor that were born outside my country!”?
Ghatanathoah
Nov 19 2019 at 1:06pm
Jones’ acceptance by immigration skeptics reminds me a lot of Fred Hoyle’s acceptance by creationists. Fred Hoyle was not a creationist, but he rejected Earth-based abiogenesis in favor of panspermia. Creationists deceptively cited the part of Hoyle’s research that criticized Earth-based abiogenesis, but ignored his arguments for panspermia. They tried to make it look like Hoyle was arguing against evolution, and therefore supported Creationism. The science is so against Creationism that they were grasping at straws trying to find any kind of validation by mainstream scientists.
Immigration skeptics are similar to Creationists. All the science is against them, so they grasp at straws whenever a respected researcher finds anything that even vaguely sounds like it is in their favor. They are also like creationists in that they tend to attribute all sorts of malign motives to the scientists who disagree with them. They don’t understand that if immigration wasn’t beneficial, that would be the result economists found. As with creationism, popular prejudices favor immigration restriction, the reason economists find otherwise is that they are fearless truthseekers who are compelled by the evidence to support open borders.
Alexander Turok
Nov 19 2019 at 2:40pm
“the reason economists find otherwise is that they are fearless truthseekers who are compelled by the evidence to support open borders.”
LOL.
Alexander Turok
Nov 19 2019 at 2:51pm
Jones is arguing against open borders here. You can say that his data and arguments logically lead to open borders, but you can’t argue that we are “deceptively citing him.”
Floccina
Nov 19 2019 at 1:07pm
And anyone with a college degree from a reasonably rigorous college from any country.
anonymous
Nov 19 2019 at 3:46pm
I don’t understand this constant motte and bailey tactic by Bryan. He argues for open borders and then if anyone argues against, he says “well you still think immigration generally is good”. So what?
Christophe Biocca
Nov 19 2019 at 5:45pm
So in general we can substantially increase immigration quotas (of whichever group the current interlocutor thinks don’t pose a problem), and even if that’s not open borders it’s much better than current policy.
He dedicates a solid fraction of his book to keyhole solutions which are basically variations on this topic.
AMT
Nov 19 2019 at 10:52pm
It’s not at all motte and bailey to say, “this is why option A is the best, but if you still disagree, then you should at least do option B, which is far better than the status quo.”
Analogy: your high school age child is skipping school and not doing their homework. So you tell them to go to class and do their homework…Or at the very least, just go to class so they can graduate, because being a dropout is worse than a graduate with poor grades.
Brian P. Moore
Nov 19 2019 at 3:11pm
I want to know policy. How close are your and Garett’s guesses on what amount of yearly immigration, constrained by current popular political beliefs, would be optimal? If we’re currently at 1M, you think we should be at 10M and he thinks 3M, and you both think political or practical constraints limit us to 2M, then it seems like you guys should be out convincing people of the 2M number, rather than debating whether you should hypothetically be fighting for 10 instead of 3. I bought Open Borders and liked it – and the best part is that it was designed to convince regular people to support increases in immigration with common sense arguments. I’d like to see more of that, from both of you!
JFA
Nov 19 2019 at 3:12pm
Quibble (minor or major I don’t know): “The Industrial Revolution didn’t just benefit factory owners; it benefited everyone who consumed the mountains of new products.” If I recall correctly, the advent of the Industrial Revolution seems was concomitant reductions in real wages and life expectancy. The decrease in life expectancy was probably driven by people crowding into nasty cities, and there is some debate over whether the decrease in wages is actually just an example of Simpson’s paradox. That being said, the Industrial Revolution (starting between 1750 and 1800) didn’t seem to lead to great things until the late 1800s. Not quite the immediate boon to society that is implied in the post above.
anonymous
Nov 19 2019 at 3:53pm
Would the entire world moving to the U.S. be similar to a one-world-government where the global population elects representatives who control policy for all countries?
anonymous
Nov 19 2019 at 4:08pm
If you run these same numbers for South Africa, don’t they come out much higher? I.e. global GDP would be much higher with the entire global population in South Africa than in the U.S.?
Alexander Turok
Nov 19 2019 at 5:13pm
I’m anti-open borders and would say yes to both questions. A 14% wage premium for natives seems far too low. The premium for English-language skills alone should be around that ballpark. But then, the assumption that if the world decamped for America they’d have ~50% of America’s productivity is also pretty unbelievable.
Mark Z
Nov 19 2019 at 9:54pm
I think Jones may actually be underselling the theoretical argument here. First, it’s not really growing population eo ipso that increases productivity, but density, or ‘interaction density’ (the range of people one can interact with). For interactions like the exchange of ideas, an increase in population pretty much automatically increases the range of people one can interact with, since technology has made distance irrelevant to communication. However, for some types of interaction, such as employment, distance still matters. Hence why urbanization boosts productivity: not just because there are more people, but because more of them are closer together.
Now, suppose 1) that the productivity benefits from increasing density (the number of people one can interact with, e.g., work for, hire, converse with, etc.) are heavily skewed toward high IQ people. Very smart people being better able to find and interact with each other increases their (and everyone else’s) productivity far more than unintelligent people being better able to find and interact with each other; and 2) what kind of people you interact with is fairly random (reflective of the composition of the population where you live). Given these assumptions, if the population (and therefore density) of a hypothetical polity (city, country, whatever) doubles, but the increase is entirely from low-IQ immigration, and the intelligent fraction of the population declines, which would reduce the rate (potentially the total number) of interactions between smart people, in turn reducing the productivity of smart people, and potentially the average productivity in general if smart people are responsible for a sufficiently disproportional amount of production.
The weak point in this argument is assumption 2. If smart people can efficiently seek and find each other, then their % of the population doesn’t matter. But to the extent that there is search inefficiency, given assumption 1, this effect becomes more plausible. An extreme example: an employer that has the necessary resources and environment to make an extremely productive innovation, should they manage hire a brilliant scientist. Under ‘perfect search efficiency,’ the employer can easily recognize which applicants meet this standard (and the applicant easily recognizes that the employer meets the necessary standards for him to be most productive). However, if the searching/matching isn’t very efficient, then the probability of the employer finding the employee they need is largely dependent on the fraction of people in the population who meet the necessary criteria; if that fraction declines, even if the number of suitable people stays the same or even goes up but by a smaller % than the overall population, the probability of an optimal match employer-employee may go down. This wouldn’t just reduce the income and ‘social surplus’ of intelligent people, but may also (if a highly intelligent/productive minority is responsible for a very large share of overall production) reduce overall productivity, as the decline in productivity in intelligent people may dwarf the modest increase the density increase has on the productivity of unintelligent people.
This isn’t peculiar to intelligence, but could apply to any productivity-associated characteristic, like conscientiousness.
stuart
Nov 20 2019 at 2:14am
Can you provide a link for the claim that white South Africans have higher IQs than black South Africans?
Jonathan S
Nov 23 2019 at 2:20am
Here’s one link:
https://www.news-medical.net/news/2005/04/26/Race-differences-in-average-IQ-are-largely-genetic.aspx
“Racial Admixture Studies. Black children with lighter skin, for example, average higher IQ scores. In South Africa, the IQ of the mixed-race “Colored” population averages 85, intermediate to the African 70 and White 100.”
Charles Lumia
Nov 20 2019 at 7:06am
First, who cares about “global production?” That’s a nonsense term that is only trying to make the nightmare scenario sound like a positive.
Nobody cares if “global production” doubles if everyone they know is worse off because of it.
Second, there would be turmoil in such a scenario. There’s turmoil with the immigration levels that we have now. To argue that increasing the numbers because of “global production” would be a good thing is nonsense that completely ignores reality.
Swami
Nov 20 2019 at 10:43am
Institutions and culture are extremely important along with the average personal characteristics of the population (intelligence, patience, trustworthiness and so on).
But the thing is, institutions, culture and population characteristics interact. The exact same institution (say democracy or traffic laws) can work completely different in a different culture (say a culture of clannishness vs individualism) or with markedly different population characteristics. And GDP is a factor of these institutions, these cultures and these population characteristics
In a world of gradual immigration, the effects of this are mitigated, as people tend to Gradually adopt the dominant cultural and institutional norms. However, Brian has not argued for immigration, he is specifically arguing for open borders, which would allow hundreds of millions of people — the poorest, with the worst institutional, cultural and personal character — to immediately become residents. These people could not afford homes in California. They will live in tents and cardboard boxes in the canyons and defecate right outside their “house” (as is expected and normal in their culture and their home country).
They will immediately go to the emergency room for free medical care paid by tax paying citizens. They will send their kids to the schools of people who actually live in houses and speak English. This will effectively destroy the school, as everything gets dumbed down to the massive influx of immigrant poor. The immigrant kids will feel like second class citizens and form gangs for support and self esteem. The non immigrant kids will respond in kind.
The politicians will pander to these new potential voters, as will Hollywood and the media. Every breaking news story will be about the rising inequality, and how cops and non immigrants are “racists” against the new folks. Elite universities will raise standards for native born Americans and set quotas for and immigrant pops, making it harder and more expensive to get native children in the best schools.
The native population, worried about the reverse discrimination against themselves,  the rampant disease and hepatitis infecting their streets and water, their escalating taxes and so on will respond by fighting back politically against anyone who supported this fiasco. If the struggle ends with the nativists winning, there will probably be more backlash against legal immigration than ever. If, God forbid, the Open Borders camp wins, then the institutions and culture of the hosting country will self implode. GDP won’t go to some kind of average between the two, it will collapse as trust decays, taxes and deficits balloon, institutions fail.
Absent the engine of prosperity in the developing countries with good institutions, culture and personal characteristics, worldwide GDP will crash, ending innovation, technology advance, and undermining free markets and representative government. This  will lead to war, global poverty, and unimaginable numbers of premature death.
Open borders is the worst idea I have ever heard.
XVO
Nov 20 2019 at 11:51am
This is maddening, you use South Africa as a pro open borders case but ignore the dysfunction caused. South Africa is actually a great example of why Garrett is right. The crime rate is high, they can’t keep the electricity on, many of their politicians support genocide and communism because they can use it to fool their low IQ constituents.
I actually agree with you that personal IQ is more important than national IQ but I think Garrett’s right about open borders. Look at living conditions in South Africa. What about Zimbabwe or Venezuela? Nations that have fully embraced their low IQ society.
Niko Davor
Dec 4 2019 at 9:42am
This is the first time I’ve seen Caplan embrace the analogy of ending apartheid in South Africa to increased immigration or open borders.
For debate purposes, this is ideal, that there is a very measurable real world scenario accepted by both sides of the ideological immigration debate. It’s easier to make convincing and quantified arguments based on a real and measurable event in the recent past, rather than hypothetical events in the future.
I’m sure nationalists and restrictionists exaggerate the costs experienced by white people as a result of the end of apartheid. However, I’m suspicious that Caplan also exaggerates his case, and skeptical of his argument that white people experienced zero income loss or other measurable costs.
I’d request debates or more convincing arguments on this issue.
Comments are closed.