Scott Sumner  

Can we have confidence in our opinions on immigration?

A Eugenic Experiment... Friday Night Video: Walter Blo...

Bryan Caplan has a post that discusses a hypothetical eugenic regime:

Imagine a Eugenic America where citizens who earn less than median income are forbidden to have children. Enforcement isn't perfect, so 5% of all kids born are "illegals." Over time, this leads to a substantial stock of people who weren't supposed to be born in the first place.

Pundits have the predictable range of positions on eugenic policy. Liberals demand amnesty for the current stock of illegals, and pledge stricter enforcement of eugenics in the future. Conservatives oppose amnesty - partly because they don't want to reward law-breaking, and partly because they don't trust liberals to help them strictly enforce eugenics laws. "Think-outside-the-box" thinkers occasionally chime in, "Fertility policy should be skill-based! Letting talented low-income people breed is good for America."

As this morally blind debate rages on, a libertarian arrives on the scene. He vocally proposes "Open Breeding." Abolish eugenics laws, and let any woman who wants a baby have a baby. Mainstream reactions are diverse, but uniformly negative.

Some commenters objected that birth restrictions are very different from immigration restrictions. If I wanted to get cute I'd point out that Bryan never claimed the two cases were similar, indeed he never mentioned immigration in the post. It was the commenters who drew that analogy. I wonder why?

Now of course Bryan probably did think this was an interesting analogy to the immigration debate. And commenters are quite correct that it's possible to argue for immigration restrictions and against birth restrictions. The two cases are not identical. But I think that misses the more interesting point. As I read Caplan, he's not really making a good argument for open borders; he's making a good argument that "we" should not trust our opinions on the open border question. By "we" I mean the 99.9% of people who have a sort of visceral negative reaction to the idea of unlimited immigration.

One can certainly imagine a society where respectable opinion believed in eugenics. Indeed in some respects that society existed 100 years ago (although actual policies were milder than in Bryan's example.) While Bryan's analogy didn't convince me that open borders are a good idea he did change my Bayesian prior on the issue. That's because I now realize that my gut instincts are not reliable. He's completely right that respectable opinion in a eugenic regime would have scoffed at Caplanesque arguments for a 100% open birth policy. That doesn't mean he's right on immigration (or indeed even eugenics), but it does suggest we should be skeptical of claims that open border proponents are somehow "loony," merely because most respectable opinion recoils from the notion of open borders. Bryan showed why respectable opinion would not be reliable in an area where many of the very same "gut instincts" come into play (such as the fear of "our civilized society being overwhelmed by barbaric hoards." My words not Bryan's.)

BTW, there is one country that does view children born in violation of government regulations as "illegals."

IN HER parents' bare brick-built shack in southern Beijing, Li Xue sifts through piles of court verdicts, petitions and other papers that record her family's struggle for most of the 20 years of her life to secure a simple document: a household registration certificate, the basic building block of official identity in China. Because she was born in violation of China's one-child-per-couple policy, local officials will not give her one. As a result she could not go to school. She now cannot get a job, nor get married, nor even buy a train or plane ticket. Despite recent moves to relax family-planning rules, the ordeal for Ms Li (pictured) is still far from over.
Read the whole thing. I've actually met academics that favor China's one child policy. I'm pretty sure they'd be horrified by this story. They'd say it's unfair to punish the innocent child for the sins of their parents. But is it really possible to have a clean, antiseptic one child policy that doesn't punish the children?

Suppose that the Chinese government had instead relied solely on a monetary fine imposed on the parent who violated the one child policy. How do you enforce that? Most peasants are too poor to pay a substantial fine. Yes, you could take away all their wealth and push them into abject poverty. But in that case you'd still be punishing the child. They would not eat well. The parents could not give them medical care; they could not afford to send them to school.

Now ask yourself how many of those academics that supported the one child policy actually thought through what would happen to the millions of children born in violation to that policy? I'd guess not very many. Now let's consider immigration restrictions. Is there a clean, antiseptic way to keep out illegal immigrants?

Bryan has not convinced me that 100% open borders are clearly the way to go. But he has convinced me that my objections to his arguments are not as reliable as I might have assumed. My reservations about open borders are actually pretty similar to the reservations that people in a eugenics society would have had to a proposal for an open birth policy.

Yes, the two cases might be different enough that the analogy doesn't hold. Maybe open births are good and open borders are bad. But the fact that superficially similar arguments against birth restrictions would have been rejected out of hand by a eugenic culture should, at the very least, make us do a bit of soul-searching.

Let's face it, most people oppose open borders at the gut level, and then they search for logical reasons to support the position that had already formed in their reptilian brain. It's all about the prospect of Kolkata on the Hudson, or Kinshasa on the Hudson. I consider myself to be less xenophobic than the average person, but even I recoil slightly at the prospect of the messy third world moving en masse to the United States, so I can understand that fear. But Bryan keeps showing us that our gut instincts on immigration might well be wrong.

Thought experiment: Suppose that in an alternative world eugenics was adopted in the year 1000. By the year 2014 the average IQ in the world was about 115 (using our current scale of 100 being average in our world.) Also suppose that there was no abject poverty, and no war. And suppose I presented a paper saying that society made a mistake in the year 1000, that eugenics should not have been adopted. A professor objects that my alternative scenario would have led to a world with billions of people with IQs in the 70s, 80s and 90s, instead of less than one billion. An alternative world with war and genocide and billions living in abject poverty. A murder rate 5 times higher. For every 10 Einsteins, there would only be one. How would I respond to that professor?

My point is not that eugenics would have been the right way to go in the year 1000. I don't believe that. Rather it is that when something has become the well-established status quo, it's really, really hard to argue for something radically different. Even if you are right.

Comments and Sharing

COMMENTS (21 to date)
Tom West writes:

A thought provoking response to a thought provoking post. Thanks.

Forcing me to evaluate my priors (and in fact recognize what are priors vs. obvious fact) is what keeps me coming back here.

David R. Henderson writes:

Beautiful piece, Scott.

Josiah writes:

[Bryan] never mentioned immigration in the post. It was the commenters who drew that analogy. I wonder why?

Because Open Borders is a subject Bryan harps on constantly on this blog.

Algernon writes:

There is a significant overlap between those who support China's one child policy and many of the funders and founders of groups that support tighter U.S. immigration restrictions.

Edward writes:

[Comment removed pending confirmation of email address and for rudeness. Email the to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

foobarista writes:

This is why I hate argument by analogy. They end up morphing into attacks or defenses on the analogies rather than a discussion of the actual topic of discussion. And it rarely "works" to change anyone's mind.

Jeff writes:

To be fair to our reptile brains, it should be noted that merely having internally open migration has, over the last fifty years, already brought a little piece of Kinshasa to Detroit, Baltimore, Oakland, St. Louis, and how many other cities to greater or lesser degrees. Forgive me if Bryan's clever extended metaphors don't make me want to stop believing my own lying eyes.

johnleemk writes:

Seems like it worked well enough to change Scott's mind, foobarista. (Obviously not "change" as in convert him, but it certainly convinced him to update his Bayesian priors.) If I were to bet, I'd wager people like Scott (folks who are already moderately pro-immigration and/or with academic economics backgrounds) are the primary intended audience of much of Bryan's posts here, rather than a hardline anti-open borders set.

Handle writes:

Scott Sumner said, "Is there a clean, antiseptic way to keep out illegal immigrants?"

It is unreasonable to call for a standard of perfection.

Is there a clean, antiseptic way to enforce any law against anybody with a dependent? Clearly not. But we still enforce laws against people with dependents. So what's the point of your rhetorical question?

foobarista writes:

Yes, it can "work", but lots of non-logical arguments can be persuasive.

Argument by analogy is typically a rhetorical argument, not a logical one, especially when the analogy one is introducing is unpleasant (if you are arguing against the original point) or pleasant (if you're arguing in favor of it).

I saw far too much of this in China, where arguments about Chinese nationalism were defended extensively using analogies to family and parents. It doesn't help that the very word for nation in Chinese is literally "family country", but all such arguments ended up dissolving into arguments about the analogy, with the main discussion abandoned.

Scott Sumner writes:

Thanks Tom and David.

Josiah, That's why I said "if I wanted to get cute. . . ."

Algernon, Interesting.

Handle, You said:

"It is unreasonable to call for a standard of perfection."

I agree with that. On the other hand I have to admit that until I read the story in The Economist I'd never given a moment's thought to what happened to kids born in defiance of the one child policy. My point is that it's easy for comfortable people like me to block out all that human suffering.

But you are right; the costs are not definitive, just something to consider.

foobarista, As I said, I don't consider the analogy itself to be something that will change people's minds. Rather it shows how unreliable our instincts are. Suppose I had lived my whole life in a world that had practiced eugenics for 1000 years. What would I think of the practice? It's hard for me to answer that question.

My views on immigration are somewhat wishy-washy. I think there is a strong argument for open borders on utilitarian grounds, (albeit some doubt resulting from differences in average and marginal effects.) At the same time I wonder if it's asking too much of Americans. Would they have to sacrifice too much? Is it reasonable to ask that much of the electorate? For me the easier question is more immigration at the margin. Raising the annual intake from 1 million a year to 2 or 3 million a year is a no-brainer.

AbsoluteZero writes:


First, thanks for the article. Analogies are just that, analogies. I don't think they're meant to be perfect in any way. But, some are more valid or useful than others. In this case, I think it is very useful and thought provoking.

If you don't mind I would like to ask a question. I don't mean anything negative. You said "... until I read the story in The Economist I'd never given a moment's thought ...". Do you remember when you first learned about the so-called one child policy, and if you tried to find out more information about it, what it is exactly, how it's implemented, and so on.

I ask because many people outside China, perhaps most, have very strong opinions about this, yet it seems when asked, they know almost next to nothing about the policy. Some people believe things that are wildly untrue. I find this rather interesting and, in a way, important.

Jim Glass writes:

Via the Flynn Effect IQ has purportedly increased by as much as 30 points in the advanced nations over the last 100 years.

So aliens breeding us for 1,000 years could've raised our average IQ by a lot more than 15 it seems. If our IQs on average were say 120 or 150 points higher than today, with all corresponding benefits, looking backward we might be grateful indeed for being the beneficiaries of such breeding.

(Thirty points is quite a lot by itself, two full standard deviations -- but having just read a couple histories of World War One, I believe it.)

John Smith writes:

Seems to me that there can be a reasonably neat and sterile manner. You do DNA testing for cases where there is an additional unauthorised child. The parent will then be shot. If both the men and the women are parents of this unauthorised child (as seems likely), the women will be shot. The men will carry on working in the outside world and leaves the kid with his parents.

Basically, much like they do it in China, except with even more brutality.

Scott Sumner writes:

Absolutezero, Hre are a few thoughts:

1. I've always been strongly opposed to the policy.

2. I was aware that it has some loopholes, the actual birthrate in China's about 1.5 per woman, I believe. It doesn't apply to minorities, and families where both parents were only children. Now (I seem to recall) it doesn't apply if only one parent is an only child.

3. I was aware of severe financial penalties for violators.

4. I had never given any thought to the fate of children who were born in violation of the policy. Here Chinese policy is even more barbaric than I anticipated, although of course I'm aware of the general level of human rights in China.

Jim, Well obviously I picked one standard deviation out of a hat. Perhaps an even bigger increase would have occurred, but I didn't want to get distracted on a debate over that issue.

However I'm a bit skeptical of the 30 point increase, although I certain know about the Flynn affect. Are we really to believe the average person was an IQ of 70 back in the old days. Perhaps, but when I read very old novels the society described doesn't seem THAT low in IQ. Wouldn't that imply half the population was close to Downs syndrome levels? I wonder if part of the story is that we are just better at taking tests, because we do it more often. Of course maybe that's what IQ measures, skill at taking test, and not the "intellectual potential" than many seem to assume. I have an open mind on the issue. Of course this is not to deny that IQ has increased, which is clearly true.

Does anyone understand John's comment, or is it just my low IQ?

Foobarista writes:

As for the one-child policy in China, it's the Chinese government's equivalent of the US War on Drugs: it's a modestly popular policy with many, vehemently opposed by some, and most importantly is the basis of a vast militarized bureaucracy that passionately believes in its mission.

libertaer writes:

I think the eugenics analogy is spot on. I'm pro open border, but you could think of crazy scenarios where Caplan too would be for closing the borders and for the same reason in favor of government enforced eugenics.

Lets say there is a gene mutation which makes some people depended on drinking human blood, they become vampires. Now, assume there is no technical solutions, it's a zero sum game, they live, we die, we live, they die. Would we open our borders for vampires? Don't think so. Now, if some Americans would carry a recessive version of this gene mutation, which -if they reproduce together- could create a vampire baby, we would want the government to stop them. (There actually are obligatory anaemia prevention programs in some countries.)

I think you can generalize that: if there is a reason related to genetics for not letting a type of person into the country, then there is a reason to stop such a type of person from being born in the country.

Now this won't convince nationalists. Here in Germany we had a big bestseller who openly advocated eugenics (by subsidizing high IQ women to have more children).

Non-nationlists see only two types of people, family (or friends) vs strangers. Nationalist think there is a middle category, some kind of metaphorical family. For them all Americans (or all Germans, Italians..) are brothers and sisters.

That's my core problem with "scientific" race-realists like Steve Sailer. I thinks he is right to emphasize the role of genes for structuring relationships. There are groups whose members think "me against my brothers, me and my brother against my cousin, me, my brother and my cousin against the world..." But it is crazy to go from small tribal societies (like Iraq, Afghanistan...) to modern nation states with millions of members and to think that genes explain their memberships. Here the social constructivist are right, "race" in the USA or Germany or France is a construct and a relative young construct. 300 million Americans are not your brothers and sisters in a non-metaphorical sense. Non-nationalst like Caplan just can't bring themselves to treat this statement literal while nationalist like Steve Sailer can't see that their thinking and feeling in metaphors.

Malcolm Kirkpatrick writes:

The analogy is exact. Natural increase is immigration from the future. If there's a way to search comments, you will find that I said this before.

1. The Earth's human population cannot grow without
2. The Earth's maximum possible instantaneous human population exceeds its maximum possible sustainable population.
3. The Earth's maximum possible sustainable human population leaves little room for wilderness or large terrestrial animals.
4. The Earth's human population will stop growing when either (a) the birth rate falls to meet the death rate or (b) the death rate rises to meet the birth rate.
5. The Earth's human population will stop growing as a result of (a) human agency or (b) other.
6. Human agency is (a) democratically controlled or (b) other.
7. All human behavioral traits are heritable.
8. Voluntary programs for population control selectively breed non-compliant individuals.
9. The government of a locality is the largest dealer in interpersonal violence in that locality.
10. Misery is like heat: in the absence of barriers (insulation, borders) it will flow until it is evenly distributed.
11. Value is determined by supply and demand. A world in which human life is precious is a world in which human life is scarce.

[Comments can be searched from our Search Page. Be sure to check the box to include the comments. --Econlib Ed.]

Malcolm Kirkpatrick writes:
Malcolm Kirkpatrick writes:

1. The Earth's human population cannot grow without ...
Proofreading lapse. Sorry

Don Geddis writes:

If you find Caplan's analogy convincing, the other alternative -- rather than question your intuitions about open borders -- would be to question your intuitions about eugenics. Perhaps the consistent resolution is that both closed borders and eugenics are good ideas.

That said, I think Sumner's scenario of eugenics in 1000 A.D. goes a little too far. No poverty? No war? That sounds like the reasoning fallacy, where if a person has one good feature, your intuition suggests that all their features are better. High IQ is just one part of a person. It doesn't relate to being more moral, less tribal, etc. Eugenics in 1000 might be able to posit more Einsteins today ... but don't confuse it with eliminating WW1&2. Hitler was evil, but not stupid.

Also, re: Flynn effect. We know that the score rise is too rapid and too large to be genetic. The Founding Fathers did not create the US out of a Down's Syndrome community. IQ tests are also more important and significant than merely measuring "test-taking skill". So neither of those is a good explanation of the data.

Comments for this entry have been closed
Return to top