“…Unfortunately, American jurisprudence forbids using IQ tests if the results lead to discrimination on the basis of race, color, religion, sex, or national origin.”
I get your thinking here and you may be right, but has discrimination wrt non-citizens residing in foreign countries been tried before the courts? Persons in this country—citizen or not—are afforded certain protections to be certain. But non-citizens residing in their own country? Can’t see the problem with vetting such folks via testing at the nearest US consulate. We should weigh IQ, education, language skills, occupation and the like. If this process restricts the admission of subsistence farmers from third world hell holes, then so be it.
Well, I hope so as well. The reference you posted certainly comes under the heading of a “person” who resides upon American controlled soil. But to assume this means all persons, everywhere in the world seems absurd. For example, the Court decision referenced that allowed IA children the right to attend public school via the 14th Amendment. So do we open schools in every country, or fly them in daily? ;-)
Non-discrimination is written into the 1965 Immigration and Nationality Act, which states that no one can be “discriminated against in the issuance of an immigrant visa because of the person’s race, sex, nationality, place of birth or place of residence.” That provision could be used, in conjunction with Griggs v. Duke Power, to strike down the use of an IQ test to screen prospective immigrants.
Disparate impact is not constitutionally codified. It only applies to certain statutory provisions, such as the title 7 pf the civil rights act, sec 2 of the voting rights act and the fair housing act.
Thus IQ tests can be applied to immigrants. The can argue that the enactment of IQ tests is a pretextual means of INTENTIONAL discrimination against certain demographics, but non citizens not on US soil likely have reduced rights under the equal protection clause (see Trump v Hawaii).
Also, and perhaps only of theoretical import, disparate impact doesnt ban IQ tests per se, but only tests that dont satisfy the business necessity test (and courts have taken the erroneous views that aptitude is not pf chief necessity in blue collar work and that job/task specific tests are more predictive than general IQ tests).
It would be necessary to amend the 1965 Immigration and Nationality Act, specifically the following provision:
"No person shall receive any preference or priority or be discriminated against in the issuance of an immigrant visa because of his race, sex, nationality, place of birth, or place of residence"
Plus there is the fact (which you alluded to) that genius is more than just hight IQ. Otherwise the world would be crawling in geniuses. It involves being an outlier in a number of other traits at the same time, which reduces the total number by several (many?) orders of magnitude.
It depends on how we define that word. Ideally, a genius should be able to interact with other people. Unfortunately, many are recluses, either because they feel little in common with other people or because are semi-autistic (and have trouble interacting socially).
Very interesting! I want to ask about the shame vs guilt culture difference: it sounds bold, which data do you base it on? I'd imagine something like MRI brain scans during shame- and guilt-causing stimuli and suchlike.
There have been MRI studies of empathy, which seems to be produced by the same mental mechanism that produces guilt. I discuss the various lines of evidence in my article on "The Large Society Problem." https://doi.org/10.4236/aa.2020.103012
>Of course, high intelligence has not evolved solely in the large high-trust societies of Europe and East Asia. It has evolved elsewhere, typically in small groups that trade with a much larger one while feeling no special responsibility for its wellbeing.
Very tasteful (and plausibly deniable) phrasing, good stuff.
I’m not necessarily against what you’re saying, but I wonder why in practice we don’t see anyone in favor of “open borders for whites.” Or, someone saying, “I’d be for more immigrants but only white ones.” People seem to be saying that (1) Indians are bad and (2) h1bs are bad because they’re too competitive or drive down wages. But point( 2) should also apply to white immigrants. Imagine if American had open borders for white English and welsh. Wages are lower and living expenses are high in England, so, in that scenario, we might see a wave of hungry economic white migrants. Would people be cool with that? I’d hope so but idk. I understand that, theoretically, there should be a compromise where we have more immigrants but whiter ones. Personally, I’d be open to that, but also I think that compromise would be DoA because no one is asking for it.
Is there any reason that importing east asians(with their shame culture) would present any different outcomes when intermingled with euro guilt cultures?
In general, East Asians have high cognitive empathy and low affective empathy. The average person has a good understanding of how other people feel, but this understanding doesn't involuntarily translate into feelings of compassion. There is much less emotional transfer from the observed person to the observer. This is why Confucianism places so much importance on learning to be compassionate.
If East Asians are a minority within a Western guilt culture, they will align themselves with the dominant pattern of behavior, especially if they adhere to Confucian or Christian values. If they are post-religious, and form the majority, the dominant behavioral pattern will be indifference toward the suffering of others.
In the West, libertarian conservatives often see this as a good thing. I don't.
Empathy and guilt proneness are the lubricants of a high-trust society. It's not just nicer to live in such a society; it's also more profitable. Many transactions are cost-effective only in that kind of environment. In a low-trust society, those same transactions would not be worth the bother.
I'm not sure what you mean by "eugenics." I believe we need to halt and reverse the current cognitive decline. This is a real thing, and it's happening faster than we think. But the emphasis should be on a large "basket" of mental and behavioral traits — not just cognitive ability, but also empathy, guilt proneness, rule following, and propensity for personal violence.
It would be better to identify communities with a high degree of social functionality (Mormons, Amish, conservative Jews and Catholics, etc.) and encourage them to procreate. We should stop this insane harassment of social conservatives for things that concern them alone, like how they raise their children.
It is an interesting contrast from the strong emphasis on familial ties of the early Europeans, where e.g the original populus Romanus consisted of thirty curiae [clans] headed by the patres [parents] of the gens [family]—all originally foreign to the Italian peninsula but that came and conquered it, forming a republic—each descended from a semi-divine patriarch.
These families constituted entire state, social, religious, and military units of their own; they were not just aristocrats, but were entire units with blood ties to each other and of common origin. Eventually the population swelled in foreign slaves, the commonwealth abounded in laws and corruption, patricians became hedonists unwilling to have families or think of their own; suddenly, the original populus Romanus was gone.
One could say that the direction began earlier in the dissolution of such clans, which happened at different times at different places. Maybe the Catholic Church only accelerated a trend that had already been long under way?
In Widukind's day, the Saxon tribes had such strong ties to each other, similar to the patricians, that they were notoriously rebellious to where Charlemagne spent most of his life trying to subdue them under what they thought the yoke of a foreign rule and religion.
To break the Saxons, Charlemagne had to forcibly and permanently separate families from each other so that they would never see their own kin again, destroying the bonds that bound them together with will to fight; this same tactic would be used much later against other populations with strong clan and blood ties like the many 'Native American' cultures.
If your going to have immigration at all, the only criteria that makes sense is to charge a fee.
For instance, I would propose we scrap the h1b program and replace it with $100,000 yearly fee that has to be paid for twenty years, end of which you get citizenship.
I’d be willing to allow five million such visas at any time, hoping to raise $500b a year to fund a child tax refund.
This would naturally limit immigrant selection to productive individuals using a market test rather than then some gameable metric.
I agree that a college degree is not a high enough standard. And as you stated, the numbers do not add up, but I am not sure that "genius" is really what we should be shooting for. That is an unrealistically high standard.
Plus skills are at least as important as intelligence. There are an awful lot of very intelligent people without skills that industry wants. Plus the acquisition of skills indirectly test for other characteristics, such as hard work and desire.
I do not see how we can assess for “high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following” except a lack of a criminal record, which is already a criteria.
What about a standard such as this?: My guess is that it will get us near 110 IQ, and it also focuses on skills, which are just as important as intelligence.
To keep the system as simple as possible, all legal immigrants must fulfill the following criteria:
1) Mastery of the English language (speaking, verbal comprehension, reading, and writing), and
2)One or more of the following:
a) A four-year college degree from an accredited university in an occupation that pays in the top third for bachelor’s degrees in the United States. The median earnings for workers with bachelor’s degrees are currently $67,005 per year and the 75th percentile is $106,004, so this should place the income cutoff at a little under $100,000 (in the US).
b) Demonstrated work experience in such an occupation.
c) The ability to pass skill-based tests that are necessary to be hired in such a field in the United States
If you can't assess those traits, or at least you can't practically for immigration purposes—obviously, you can assess them, as otherwise we wouldn't know certain areas *have* those traits and not others, and I remember Frost going into the experiments that established this before, so I don't see why these couldn't be applied to immigration process, but for the sake of argument if we can't—then if you agree on their importance, wouldn't the sensible options be to either:
1. Stop immigration in its modern form altogether because of this inability to measure the most important traits for compatibility.
2. Only allow immigration from geographic areas (be they particular nation states or more specific than that) whose populations are high in these traits already.
Unfortunately, both of the above are considered intolerably discriminatory to most people, out of what normally seems to be a psychological reaction or pair association than a well-thought out view.
An interesting fact that Frost has presented before is that it's dysgenic to other countries to take their smartest and most-educated populations from them. It harms immigrants' own countries and populations to import them, depriving them of their best citizens and making it impossible for them to compete for labour when the market is international. In addition to the harm to the host countries in this same way. It interferes with complex processes of nature and of evolution in ways we can't fully predict or know.
If someone is unable to appreciate the harm immigration does to their own country out of sympathy for individuals, I find they are usually more willing to consider the harm it does to other countries. E.g.: Immigration has prevented African countries like Zimbabwe from being able to employ sufficient nurses and doctors; they train them themselves, only for them to leave.
In principle, I think all peoples deserve autonomy and to be allowed to have their own culture, their own religion, to exercise their own free will and choose their own path without interference. Modern immigration will not ever allow for this; the free will of entire nations is violated in favour of the freedom of selfish individuals and employers.
I just gave you an example of how to assess the most important traits for economic success. Something like that should be done regardless of the region. We let individual immigrants come into the nation, not entire regions.
Let's say, for example, that we will only allow Western Europeans to enter the US. That still includes tens of millions of people. There are far more individual variations within regions than there are regional differences.
I would much rather have a talented African mechanical engineer than a random Western European. That person has already demonstrated success and useful skills, whereas a random Western European just has the potential for those. Therefore, we do not need to see proof of “high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following”
Social trust is made possible by high levels of empathy, guilt proneness, and rule following. These characteristics have a heritable basis, but they are independent of intelligence. Yes, they correlate with intelligence in many studies, but that is only because those studies have been done with subjects of Western European descent.
Other cultural contexts have selected for other mental and behavioral packages. In many cultures, high intelligence correlates with concern only for close relatives, and indifference toward everyone else.
So how do you assess “ high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following” for an individual?
You claim in other comment that it is difficult to assess skills without corruption, but this characteristic seems far harder to assess.
It could be done, but it would be legally impossible for the same reason that an IQ test would be legally impossible. It would essentially ban immigration from many countries, and that outcome would violate the non-discriminatory provision of the 1965 Immigration Act:
"No person shall receive any preference or priority or be discriminated against in the issuance of an immigrant visa because of his race, sex, nationality, place of birth, or place of residence"
'...we do not need to see proof of “high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following”'
This depends on priorities. You might--*might*--be right if your only interest is in 'economic success' (but measured how?) If you are interested mainly in social cohesion, your criteria will be less likely to work. Surely you can understand this.
Your categorical opposition between 'a talented African mechanical engineer than a random Western European' is disingenuous. The choice is always between two people with talent in the same domain, not a talented ____ and a random_____.
One way to 'recruit' high social-trust individuals, and to justify their recruitment, would be through a simple examination and publicisation of the trust section of World Values Survey data, in which W European(-derived) societies always come out on top.
Regarding "The choice is always between two people with talent in the same domain."
No, that is absolutely not true in the current American immigration system. Very few immigrants are allowed to enter based on their skills. The vast majority are for family reunification or refugees/asylees. That is exactly what I want to change.
Yet I dare say there is never a decision to be made between a 'random' white person and a 'talented African mechanical engineer' (your opposition, not mine). I say that, *if* we are talking about two applicants with nominally equivalent skills, the European should be admitted. Since you are determined to reduce everything to the level of the individual, how else would you choose between the two applicants in this hypothetical case--toss a coin? You talk as if individuals are random exponents of group attributes; they obviously are not.
I am not determined "to reduce everything to the level of the individual." That is how all immigration systems have to work. You cannot let in an entire nation.
As for your example, No. If they both meet the criterion, they both get in.
You only have to choose between them when we are close to the annual limit, which is currently 675,000. If by chance, those two immigrants apply when we are right at the limit, then a coin flip is fine. The loser will likely get admitted one month later. Most likely one applied before the other, so there is no need for a coin flip.
Problem solved.
I have no idea what "random exponents of group attributes" means.
The reality is that far more Europeans will meet the criteria than Africans, so it is not really a problem in the real world (as I mention in my article).
'How do you apply the World Values System survey to one individual?'
Obviously it would be impracticable to do so. You'd proceed instead on the very simple basis that central-tendency measurements are typical of most individuals belonging to the group at issue.
What you describe is the Canadian immigration system. We have a point system that prioritizes immigrants according to their educational level, their knowledge of French or English, and their relevant employment experience.
Initially, that system seemed to work. Then, increasingly, it didn't. South Asians used to have very low rates of violent crime in Canada. Now, their violent crime rates are greatly above the Canadian average. Meanwhile, the dependency ratio of immigrants has been steadily rising. And the pressure on the health care system has been disproportionately increased by immigration. Immigrants are supposed to be proficient in English or French, yet, increasingly, a form of pidgin English has become the norm.
Things went wrong for several reasons:
- Family reunification grew exponentially, especially for groups with strong kinship ties. There has thus been regression to the mean, as immigrants sponsor relatives, who then sponsor their relatives.
- Once we set up non-discrimination as a desired value, there was an inevitable push to make the system more and more non-discriminatory. Critics pointed out that the point system still favored immigrants of European descent, and this criticism led to a loosening of criteria to let in more immigrants from the Third World.
- Legal immigration facilitated illegal immigration. As Third World communities grew in size, they became havens for illegal Third World immigrants.
- Once we made a college education a criterion for admission to Canada, we incentivized the creation of college programs whose express purpose was to help people become eligible for immigration. These "degree mills" have now become a major source of income for many postsecondary institutions in Canada and abroad.
- The same has been true for all of the other immigration criteria: job offers, "clean" criminal records, attestations of language proficiency, etc. Once you create a demand for fake qualifications, businesses will emerge to meet that demand.
No, my proposal is not the Canadian points system. Under Trudeau, Canada abandoned its focus on skilled immigrants. Indeed, I specifically mention in the article that we should not implement such a system.
My proposal is actually quite different.
A few examples:
1) (probably most important) The Canadian system deliberately radically increased the number of immigrants, not mine.
2) my proposal does not allow for family reunification (unless all family members meet the criteria, which is very unlikely).
3) my proposal includes strict enforcement against illegal immigration at the border and internally
4) A college degree alone is not a criterion for admission in my system
5) Degree mills are easy to avoid with skills tests. A simple interview by an expert in the field will expose them.
6) My proposal has no points.
I never "set up non-discrimination as a desired value." Immigration policy by definition lets in one person and says no to another. That is discrimination, and that is exactly what our immigration should do. People forget what the word actually means.
Within the Indo-Canadian community, violent crime generally takes the form of extortion, rather than homicide (although homicide can be involved). Nine out of Canada's 11 leading gangsters are of Punjabi origin. https://www.staysafevancouver.com/post/indo-canadian-organized-crime
It's difficult to get statistics because most incidents of extortion go unreported: "law enforcement officials say there are likely more incidents, but that some victims aren't coming forward and instead are simply paying the extortion." https://www.cbc.ca/news/canada/extortion-attacks-south-asian-1.7134264
If the top 0.1% of engineers is only 6 or 7 points to the right (and not 10 points), we're talking about people with an IQ of 153 or more. That is still a very small population.
When I look at the charts in Wolfram (2023), I don't get the impression that engineers have a narrower IQ distribution than other occupational groups. The mean IQ of engineers also seems to be more like 112 than 110
'Even if one could find a legally acceptable way to select prospective immigrants for intelligence, the results would still be disappointing. It is not simply that the world has fewer geniuses than some of us think. There is also the need to screen for other mental traits that are no less important for the survival of a high-trust society.'
As someone familiar with American culture, I can say that America will never have a merit-based or points-based immigration system. That's because American immigration is mostly about family reunification; family is a very important concept in this country. Also, apart from family, to the extent that immigration exists, America seems to want the *low-skilled* variety only. The tacit "open border" with Mexico that everyone complains about illustrates this well. The fruit pickers, hotel maintenance workers, construction workers, and meatpacking workers are secretly in great demand and often America is prepared to look the other way, although now the new admin is expected to crack down on that. But you never had to be some kind of intelligent person to immigrate to America, quite the opposite. The poorest and most blue-collar people imaginable ("the unwashed masses") went through Ellis Island and the same is true now. In addition, as Vivek Ramaswamy correctly said (although this is a bit insulting coming from someone with an immigrant background), America values entertainment, sports, and glamor over intelligence and academics. Being smart or a nerd is looked down upon. What matters is whether you're good-looking and athletic. That's just the way American culture is.
Family reunification is a big issue only for cultures with large extended families. Americans generally have weak kinship ties, and most are opposed to family reunification and the endless chain migration it produces.
It's important to distinguish between reality and the self-serving myths generated by an oligarch-dominated media.
"Engineers have a mean IQ of about 110 (Wolfram, 2023).
The top 0.1% of Americans have IQs of 147 or more. Therefore, the top 0.1% of engineers is ten points to the right, i.e., IQs of 157 or more."
This is wrong on two levels. You assume the two factors, being a 0.1% engineer and being 0.1% IQ are independent, when the reasons IQ might correlate with high performance don't even need to be stated they are so obvious.
Second, you misunderstand how normal distributions work. Simply adding 10 points to the general population's top 0.1% IQ (147) assumes a linear shift at all levels, which isn’t how normal distributions work. This is because the extreme end of the distribution grows narrower and doesn't shift by a fixed amount.
"American college students are now just above the cognitive average, with an estimated mean IQ of 102"
This is debatably true of American college students (there's good reason to believe this estimated mean is incorrect), but is irrelevant for international students. The sorts of people who can manage to achieve high enough grades to succeed in American universities, often not in their native language, sets them a level above the national average (which includes schools all the way down to community colleges). Being an international student has already passed you through a serious qualification procedure.
"Unfortunately, American jurisprudence forbids using IQ tests if the results lead to discrimination on the basis of race, color, religion, sex, or national origin."
This is obviously not true when it comes to accepting foreign immigrants. Our current system very explicitly discriminates based on national origin when we set maximum number of immigrants from any specific country.
1. I'm assuming that American engineer IQ follows the same normal distribution as that of American IQ. If the mean IQ of American engineers is 110, the 0.1% cutoff for them should be ten points farther to the right than the 0.1% cutoff for Americans in general.
If you're having trouble thinking this through, just imagine a picture of a bell curve with its center at 100. Now superimpose the same bell curve on that picture but 10 points farther to the right. The 0.1% cutoff should likewise be 10 points farther to the right.
Now, it's possible that the IQ distribution of American engineers is different. Perhaps it's more skewed. The difference, however, would have to be considerable to change my conclusion. If you still disagree, why don't you crunch the numbers and offer your own estimate? I don't see how it could be substantially different. The total of geniuses (157+) in the entire world cannot be much more than half a million.
2. You're wrong. Domestic students consistently outperform international students.
He, Yunke, and Heather Banham. "International student academic performance: Some statistical evidence and its implications." American Journal of Business Education 2.5 (2009): 89-100. https://eric.ed.gov/?id=EJ1052845
3. In the past, American courts would rule that American principles of non-discrimination do not apply to non-citizens. Recently, there has been a trend to treat non-citizens as if they were citizens.
1. That's a poor assumption. There are simple reasons why a subset of the population would not have a normal distribution. If almost no one of <90 IQ becomes an engineer, this would also boost the average significantly, without changing the IQ of the top 0.1% of engineers by more than a point. What do you think is more likely, that there's a sharp drop in engineers below some certain threshold of IQ, or that the likelihood of someone becoming an engineer is independent of IQ?
But let me run the numbers.
If you preclude all individuals with IQs less than 97, you would get a new, heavily skewed distribution with a mean of 110. While in reality I'm sure there are many programmers with less than 97 IQ, the likelihood of one becoming a programmer probably decreases with decreasing IQ, and my hard cutoff is just a simplification to make calculation easy.
This new distribution (eliminating those with under 97 IQ would create an approximately folded normal distribution) makes it very easy to calculate the SD for a 0.1% occurrence. I get 146.35 from my calculations. Of course in reality it isn't a perfect folded normal distribution, (it would be if the cutoff for a mean 110 IQ was 100), and there's a not-insignificant amount of people in those three points between 97 and 100, so the actual 0.1% would be slightly higher. The real number would be between 148-150, without doing the much more complicated math of calculating SD for a non-standard distribution.
This is all because, while eliminating under performers can significant improve the mean, the upper tail distribution barely changes.
2. Looking more into this, the evidence is extremely limited, with that literally being the only study on the topic of international vs. domestic student performance I can find, which is extremely odd. It seems there may be a taboo on comparing these two groups (not surprising in modern academia), as basically anyone with access to a schools admin software should (in theory) be able to easily compare mean GPA, graduation rate, etc. between the two types of students in an afternoon and make a study about it.
I'll take back my previous statement on this and say I'm unsure. I think there's good reason to be skeptical of this study though, as it seems it only deals with one school, the Okanagan School of Business in British Columbia, Canada, which has a different, and far more easily exploitable international student Visa system than the US.
I'll default to the common knowledge that international students face 1/3-1/2 the acceptance rates of domestic students (no studies I could find estimating this, which may be more evidence of tabboo), especially at top universities, as positions are limited and number of applicants are much higher. More selective generally equates to higher quality, although I can imagine some reasons that this isn't as strong of an effect due to a lower quality applicant pool for international students.
3. This is fair, and potentially a future issue, but discrimination based on national origin is literally the core of the current H1-B system with its national quotas. Any new H1-B program created by the Trump administration will definitely implement national discrimination.
1. The IQ distribution of engineers may be skewed, but even a highly skewed distribution wouldn't change things much. There just aren't that many geniuses in the world, and most are already in affluent countries. It's naïve to think we can "mine" India (or any country) for large numbers of geniuses, and I question the honesty of people who make that argument. They should come clean and say their goal is not immigration of geniuses, but rather immigration of somewhat-above-average people who can do IT grunge work in Silicon Valley.
2. The problem in Canada is not so much that international students are more exploitable but rather that a larger proportion of them don't really see themselves as students. They see the IS program as a means to enter the country and gain permanent status.
3. I honestly don't know. Court rulings are difficult to predict because they hinge on political beliefs, and such beliefs can differ from one judge to another. There will probably be several court decisions before this issue is settled.
1. That’s fair. The 0.1% talent thing is kind of a ridiculous statement, as that’s inevitably going to be 1/1000 of the total number of software engineers, assuming literally all the top software engineers want to come to the US to study, and not stay in India or study in another country like the UK. Having known many international students, they are often obviously very intelligent hardworking people, and any calculation is going to be downwind of that. I still stand by my calculations that the top .1% of engineers won’t have significantly higher IQ than the top .1% of the population in general though, based off my assumptions.
2. I meant more easily exploitable by the students. You’re definitely right that most don’t treat it as coming to study but just coming to immigrate. I believe there’s no cap on the number of years you can take to finish a degree in their system (or its high like 8 years) so students take 2 credits a semester or something while working full time.
I think It'd be more than 7,000,000 because (10^9)*(100-99.993) = 7 million. 157+ is a bit less than 4 standard deviations in China, and China has like 1 billion people (order-of-magnitude estimate purposes). That being said, I get the (extremely uninformed, mind you) impression that there is a lot of focus is on getting talent from India, ignoring the fact that most of the cognitive elite of that country might already be overseas.
Moreover, the U.S. definitely lacks the ability to attract most of the cognitive elite in China.
Also, I wonder to what extent the historically high levels of endogamous stratification in many regions of the Third World, such as Nigeria or Saudi Arabia, affect the existence of a cognitive elite in those countries. Perhaps these regions have thicker tails (from both high and low castes) than one would expect from a normal distribution.
Nvm, I'm clearly innumerate. Four standard deviations is way more extreme than that. I should have checked with R instead of trusting information online...
“…Unfortunately, American jurisprudence forbids using IQ tests if the results lead to discrimination on the basis of race, color, religion, sex, or national origin.”
I get your thinking here and you may be right, but has discrimination wrt non-citizens residing in foreign countries been tried before the courts? Persons in this country—citizen or not—are afforded certain protections to be certain. But non-citizens residing in their own country? Can’t see the problem with vetting such folks via testing at the nearest US consulate. We should weigh IQ, education, language skills, occupation and the like. If this process restricts the admission of subsistence farmers from third world hell holes, then so be it.
I want to believe you, but the courts are moving in that direction, not only in the US but also in Canada and in the European Union. https://www.maniatislawoffice.com/blog/2018/08/do-non-citizens-have-constitutional-rights/
Well, I hope so as well. The reference you posted certainly comes under the heading of a “person” who resides upon American controlled soil. But to assume this means all persons, everywhere in the world seems absurd. For example, the Court decision referenced that allowed IA children the right to attend public school via the 14th Amendment. So do we open schools in every country, or fly them in daily? ;-)
Non-discrimination is written into the 1965 Immigration and Nationality Act, which states that no one can be “discriminated against in the issuance of an immigrant visa because of the person’s race, sex, nationality, place of birth or place of residence.” That provision could be used, in conjunction with Griggs v. Duke Power, to strike down the use of an IQ test to screen prospective immigrants.
Disparate impact is not constitutionally codified. It only applies to certain statutory provisions, such as the title 7 pf the civil rights act, sec 2 of the voting rights act and the fair housing act.
Thus IQ tests can be applied to immigrants. The can argue that the enactment of IQ tests is a pretextual means of INTENTIONAL discrimination against certain demographics, but non citizens not on US soil likely have reduced rights under the equal protection clause (see Trump v Hawaii).
Also, and perhaps only of theoretical import, disparate impact doesnt ban IQ tests per se, but only tests that dont satisfy the business necessity test (and courts have taken the erroneous views that aptitude is not pf chief necessity in blue collar work and that job/task specific tests are more predictive than general IQ tests).
It would be necessary to amend the 1965 Immigration and Nationality Act, specifically the following provision:
"No person shall receive any preference or priority or be discriminated against in the issuance of an immigrant visa because of his race, sex, nationality, place of birth, or place of residence"
https://immigrationhistory.org/item/hart-celler-act/
How does a ban on IQ tests follow from the quoted statutory language?
Wow! Amazing post! This has answered so many questions that I have wondered about. Thank you so much!
Plus there is the fact (which you alluded to) that genius is more than just hight IQ. Otherwise the world would be crawling in geniuses. It involves being an outlier in a number of other traits at the same time, which reduces the total number by several (many?) orders of magnitude.
It depends on how we define that word. Ideally, a genius should be able to interact with other people. Unfortunately, many are recluses, either because they feel little in common with other people or because are semi-autistic (and have trouble interacting socially).
Very interesting! I want to ask about the shame vs guilt culture difference: it sounds bold, which data do you base it on? I'd imagine something like MRI brain scans during shame- and guilt-causing stimuli and suchlike.
There have been MRI studies of empathy, which seems to be produced by the same mental mechanism that produces guilt. I discuss the various lines of evidence in my article on "The Large Society Problem." https://doi.org/10.4236/aa.2020.103012
>Of course, high intelligence has not evolved solely in the large high-trust societies of Europe and East Asia. It has evolved elsewhere, typically in small groups that trade with a much larger one while feeling no special responsibility for its wellbeing.
Very tasteful (and plausibly deniable) phrasing, good stuff.
I’m not necessarily against what you’re saying, but I wonder why in practice we don’t see anyone in favor of “open borders for whites.” Or, someone saying, “I’d be for more immigrants but only white ones.” People seem to be saying that (1) Indians are bad and (2) h1bs are bad because they’re too competitive or drive down wages. But point( 2) should also apply to white immigrants. Imagine if American had open borders for white English and welsh. Wages are lower and living expenses are high in England, so, in that scenario, we might see a wave of hungry economic white migrants. Would people be cool with that? I’d hope so but idk. I understand that, theoretically, there should be a compromise where we have more immigrants but whiter ones. Personally, I’d be open to that, but also I think that compromise would be DoA because no one is asking for it.
Is there any reason that importing east asians(with their shame culture) would present any different outcomes when intermingled with euro guilt cultures?
In general, East Asians have high cognitive empathy and low affective empathy. The average person has a good understanding of how other people feel, but this understanding doesn't involuntarily translate into feelings of compassion. There is much less emotional transfer from the observed person to the observer. This is why Confucianism places so much importance on learning to be compassionate.
If East Asians are a minority within a Western guilt culture, they will align themselves with the dominant pattern of behavior, especially if they adhere to Confucian or Christian values. If they are post-religious, and form the majority, the dominant behavioral pattern will be indifference toward the suffering of others.
In the West, libertarian conservatives often see this as a good thing. I don't.
I think it might be a good thing to the extent that east asian low empathy allows for enactment of pro eugenic policies. What are your objections?
Empathy and guilt proneness are the lubricants of a high-trust society. It's not just nicer to live in such a society; it's also more profitable. Many transactions are cost-effective only in that kind of environment. In a low-trust society, those same transactions would not be worth the bother.
I'm not sure what you mean by "eugenics." I believe we need to halt and reverse the current cognitive decline. This is a real thing, and it's happening faster than we think. But the emphasis should be on a large "basket" of mental and behavioral traits — not just cognitive ability, but also empathy, guilt proneness, rule following, and propensity for personal violence.
It would be better to identify communities with a high degree of social functionality (Mormons, Amish, conservative Jews and Catholics, etc.) and encourage them to procreate. We should stop this insane harassment of social conservatives for things that concern them alone, like how they raise their children.
Europe went in a different direction in the Middle Ages when the Catholic Church decided to break up the clans by forbidding cousin marriage.
It became ingrained in the culture and wasn’t reversed after the Protestant Reformation. Nowhere did this as far as I know.
Thus family ties became less important than they are in other societies (where they’re often extremely important).
It is an interesting contrast from the strong emphasis on familial ties of the early Europeans, where e.g the original populus Romanus consisted of thirty curiae [clans] headed by the patres [parents] of the gens [family]—all originally foreign to the Italian peninsula but that came and conquered it, forming a republic—each descended from a semi-divine patriarch.
These families constituted entire state, social, religious, and military units of their own; they were not just aristocrats, but were entire units with blood ties to each other and of common origin. Eventually the population swelled in foreign slaves, the commonwealth abounded in laws and corruption, patricians became hedonists unwilling to have families or think of their own; suddenly, the original populus Romanus was gone.
One could say that the direction began earlier in the dissolution of such clans, which happened at different times at different places. Maybe the Catholic Church only accelerated a trend that had already been long under way?
In Widukind's day, the Saxon tribes had such strong ties to each other, similar to the patricians, that they were notoriously rebellious to where Charlemagne spent most of his life trying to subdue them under what they thought the yoke of a foreign rule and religion.
To break the Saxons, Charlemagne had to forcibly and permanently separate families from each other so that they would never see their own kin again, destroying the bonds that bound them together with will to fight; this same tactic would be used much later against other populations with strong clan and blood ties like the many 'Native American' cultures.
If your going to have immigration at all, the only criteria that makes sense is to charge a fee.
For instance, I would propose we scrap the h1b program and replace it with $100,000 yearly fee that has to be paid for twenty years, end of which you get citizenship.
I’d be willing to allow five million such visas at any time, hoping to raise $500b a year to fund a child tax refund.
This would naturally limit immigrant selection to productive individuals using a market test rather than then some gameable metric.
I agree that a college degree is not a high enough standard. And as you stated, the numbers do not add up, but I am not sure that "genius" is really what we should be shooting for. That is an unrealistically high standard.
Plus skills are at least as important as intelligence. There are an awful lot of very intelligent people without skills that industry wants. Plus the acquisition of skills indirectly test for other characteristics, such as hard work and desire.
I do not see how we can assess for “high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following” except a lack of a criminal record, which is already a criteria.
What about a standard such as this?: My guess is that it will get us near 110 IQ, and it also focuses on skills, which are just as important as intelligence.
To keep the system as simple as possible, all legal immigrants must fulfill the following criteria:
1) Mastery of the English language (speaking, verbal comprehension, reading, and writing), and
2)One or more of the following:
a) A four-year college degree from an accredited university in an occupation that pays in the top third for bachelor’s degrees in the United States. The median earnings for workers with bachelor’s degrees are currently $67,005 per year and the 75th percentile is $106,004, so this should place the income cutoff at a little under $100,000 (in the US).
b) Demonstrated work experience in such an occupation.
c) The ability to pass skill-based tests that are necessary to be hired in such a field in the United States
I go into more detail here:
https://frompovertytoprogress.substack.com/p/we-need-a-skills-based-immigration
If you can't assess those traits, or at least you can't practically for immigration purposes—obviously, you can assess them, as otherwise we wouldn't know certain areas *have* those traits and not others, and I remember Frost going into the experiments that established this before, so I don't see why these couldn't be applied to immigration process, but for the sake of argument if we can't—then if you agree on their importance, wouldn't the sensible options be to either:
1. Stop immigration in its modern form altogether because of this inability to measure the most important traits for compatibility.
2. Only allow immigration from geographic areas (be they particular nation states or more specific than that) whose populations are high in these traits already.
Unfortunately, both of the above are considered intolerably discriminatory to most people, out of what normally seems to be a psychological reaction or pair association than a well-thought out view.
An interesting fact that Frost has presented before is that it's dysgenic to other countries to take their smartest and most-educated populations from them. It harms immigrants' own countries and populations to import them, depriving them of their best citizens and making it impossible for them to compete for labour when the market is international. In addition to the harm to the host countries in this same way. It interferes with complex processes of nature and of evolution in ways we can't fully predict or know.
If someone is unable to appreciate the harm immigration does to their own country out of sympathy for individuals, I find they are usually more willing to consider the harm it does to other countries. E.g.: Immigration has prevented African countries like Zimbabwe from being able to employ sufficient nurses and doctors; they train them themselves, only for them to leave.
In principle, I think all peoples deserve autonomy and to be allowed to have their own culture, their own religion, to exercise their own free will and choose their own path without interference. Modern immigration will not ever allow for this; the free will of entire nations is violated in favour of the freedom of selfish individuals and employers.
You are not really responding to my proposal.
I just gave you an example of how to assess the most important traits for economic success. Something like that should be done regardless of the region. We let individual immigrants come into the nation, not entire regions.
Let's say, for example, that we will only allow Western Europeans to enter the US. That still includes tens of millions of people. There are far more individual variations within regions than there are regional differences.
I would much rather have a talented African mechanical engineer than a random Western European. That person has already demonstrated success and useful skills, whereas a random Western European just has the potential for those. Therefore, we do not need to see proof of “high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following”
Social trust is made possible by high levels of empathy, guilt proneness, and rule following. These characteristics have a heritable basis, but they are independent of intelligence. Yes, they correlate with intelligence in many studies, but that is only because those studies have been done with subjects of Western European descent.
Other cultural contexts have selected for other mental and behavioral packages. In many cultures, high intelligence correlates with concern only for close relatives, and indifference toward everyone else.
So how do you assess “ high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following” for an individual?
You claim in other comment that it is difficult to assess skills without corruption, but this characteristic seems far harder to assess.
It could be done, but it would be legally impossible for the same reason that an IQ test would be legally impossible. It would essentially ban immigration from many countries, and that outcome would violate the non-discriminatory provision of the 1965 Immigration Act:
"No person shall receive any preference or priority or be discriminated against in the issuance of an immigrant visa because of his race, sex, nationality, place of birth, or place of residence"
https://immigrationhistory.org/item/hart-celler-act/
But how could it be done?
Just stating that it can be done is not persuasive.
And laws can be amended.
'...we do not need to see proof of “high level of social trust — a result of equally high levels of empathy, guilt proneness, and rule following”'
This depends on priorities. You might--*might*--be right if your only interest is in 'economic success' (but measured how?) If you are interested mainly in social cohesion, your criteria will be less likely to work. Surely you can understand this.
Your categorical opposition between 'a talented African mechanical engineer than a random Western European' is disingenuous. The choice is always between two people with talent in the same domain, not a talented ____ and a random_____.
One way to 'recruit' high social-trust individuals, and to justify their recruitment, would be through a simple examination and publicisation of the trust section of World Values Survey data, in which W European(-derived) societies always come out on top.
Regarding "The choice is always between two people with talent in the same domain."
No, that is absolutely not true in the current American immigration system. Very few immigrants are allowed to enter based on their skills. The vast majority are for family reunification or refugees/asylees. That is exactly what I want to change.
Yet I dare say there is never a decision to be made between a 'random' white person and a 'talented African mechanical engineer' (your opposition, not mine). I say that, *if* we are talking about two applicants with nominally equivalent skills, the European should be admitted. Since you are determined to reduce everything to the level of the individual, how else would you choose between the two applicants in this hypothetical case--toss a coin? You talk as if individuals are random exponents of group attributes; they obviously are not.
Something tells me that you did not even read the article in my comment that triggered this discussion...
I am not determined "to reduce everything to the level of the individual." That is how all immigration systems have to work. You cannot let in an entire nation.
As for your example, No. If they both meet the criterion, they both get in.
You only have to choose between them when we are close to the annual limit, which is currently 675,000. If by chance, those two immigrants apply when we are right at the limit, then a coin flip is fine. The loser will likely get admitted one month later. Most likely one applied before the other, so there is no need for a coin flip.
Problem solved.
I have no idea what "random exponents of group attributes" means.
The reality is that far more Europeans will meet the criteria than Africans, so it is not really a problem in the real world (as I mention in my article).
How do you apply the World Values System survey to one individual?
There is substantial variation among individuals within each region.
Yes, I am less concerned about social cohesion, particularly given that highly skilled workers do not tend to threaten it.
'How do you apply the World Values System survey to one individual?'
Obviously it would be impracticable to do so. You'd proceed instead on the very simple basis that central-tendency measurements are typical of most individuals belonging to the group at issue.
'Yes, I am less concerned about social cohesion.'
So I gathered
So you are not even going to try to assess an individual on what you apparently regard as the most important criteria of admission?
Seems odd.
What you describe is the Canadian immigration system. We have a point system that prioritizes immigrants according to their educational level, their knowledge of French or English, and their relevant employment experience.
Initially, that system seemed to work. Then, increasingly, it didn't. South Asians used to have very low rates of violent crime in Canada. Now, their violent crime rates are greatly above the Canadian average. Meanwhile, the dependency ratio of immigrants has been steadily rising. And the pressure on the health care system has been disproportionately increased by immigration. Immigrants are supposed to be proficient in English or French, yet, increasingly, a form of pidgin English has become the norm.
Things went wrong for several reasons:
- Family reunification grew exponentially, especially for groups with strong kinship ties. There has thus been regression to the mean, as immigrants sponsor relatives, who then sponsor their relatives.
- Once we set up non-discrimination as a desired value, there was an inevitable push to make the system more and more non-discriminatory. Critics pointed out that the point system still favored immigrants of European descent, and this criticism led to a loosening of criteria to let in more immigrants from the Third World.
- Legal immigration facilitated illegal immigration. As Third World communities grew in size, they became havens for illegal Third World immigrants.
- Once we made a college education a criterion for admission to Canada, we incentivized the creation of college programs whose express purpose was to help people become eligible for immigration. These "degree mills" have now become a major source of income for many postsecondary institutions in Canada and abroad.
- The same has been true for all of the other immigration criteria: job offers, "clean" criminal records, attestations of language proficiency, etc. Once you create a demand for fake qualifications, businesses will emerge to meet that demand.
No, my proposal is not the Canadian points system. Under Trudeau, Canada abandoned its focus on skilled immigrants. Indeed, I specifically mention in the article that we should not implement such a system.
My proposal is actually quite different.
A few examples:
1) (probably most important) The Canadian system deliberately radically increased the number of immigrants, not mine.
2) my proposal does not allow for family reunification (unless all family members meet the criteria, which is very unlikely).
3) my proposal includes strict enforcement against illegal immigration at the border and internally
4) A college degree alone is not a criterion for admission in my system
5) Degree mills are easy to avoid with skills tests. A simple interview by an expert in the field will expose them.
6) My proposal has no points.
I never "set up non-discrimination as a desired value." Immigration policy by definition lets in one person and says no to another. That is discrimination, and that is exactly what our immigration should do. People forget what the word actually means.
All immigration systems discriminate.
Is violent crime rate of South Asians really greatly above the Canadian average? Data from 2019-2023 shows that South Asian males, as well as West Asian and Arab males, have homicide rates well below average: https://www150.statcan.gc.ca/t1/tbl1/en/tv.action?pid=3510020701&pickMembers%5B0%5D=1.1&pickMembers%5B1%5D=2.2&pickMembers%5B2%5D=4.3&cubeTimeFrame.startYear=2019&cubeTimeFrame.endYear=2023&referencePeriods=20190101%2C20230101
I'd be interested to know if there's data showing the contrary.
Within the Indo-Canadian community, violent crime generally takes the form of extortion, rather than homicide (although homicide can be involved). Nine out of Canada's 11 leading gangsters are of Punjabi origin. https://www.staysafevancouver.com/post/indo-canadian-organized-crime
It's difficult to get statistics because most incidents of extortion go unreported: "law enforcement officials say there are likely more incidents, but that some victims aren't coming forward and instead are simply paying the extortion." https://www.cbc.ca/news/canada/extortion-attacks-south-asian-1.7134264
In Canada, the rate of extortion has increased 7-fold since 2010. https://www150.statcan.gc.ca/n1/daily-quotidien/240725/cg-b004-eng.htm
"The top 0.1% of Americans have IQs of 147 or more. Therefore, the top 0.1% of engineers is ten points to the right, i.e., IQs of 157 or more."
You assumed implicitly that the SD of engineers is also 15. This is not true, it's more like 10.
If the top 0.1% of engineers is only 6 or 7 points to the right (and not 10 points), we're talking about people with an IQ of 153 or more. That is still a very small population.
When I look at the charts in Wolfram (2023), I don't get the impression that engineers have a narrower IQ distribution than other occupational groups. The mean IQ of engineers also seems to be more like 112 than 110
https://doi.org/10.1016/j.intell.2023.101755
Necessarily, selected subsets of the population have a smaller SD. This effect has long been noted in the literature. It's section 4.4 in that paper.
'Even if one could find a legally acceptable way to select prospective immigrants for intelligence, the results would still be disappointing. It is not simply that the world has fewer geniuses than some of us think. There is also the need to screen for other mental traits that are no less important for the survival of a high-trust society.'
Exactly so
As someone familiar with American culture, I can say that America will never have a merit-based or points-based immigration system. That's because American immigration is mostly about family reunification; family is a very important concept in this country. Also, apart from family, to the extent that immigration exists, America seems to want the *low-skilled* variety only. The tacit "open border" with Mexico that everyone complains about illustrates this well. The fruit pickers, hotel maintenance workers, construction workers, and meatpacking workers are secretly in great demand and often America is prepared to look the other way, although now the new admin is expected to crack down on that. But you never had to be some kind of intelligent person to immigrate to America, quite the opposite. The poorest and most blue-collar people imaginable ("the unwashed masses") went through Ellis Island and the same is true now. In addition, as Vivek Ramaswamy correctly said (although this is a bit insulting coming from someone with an immigrant background), America values entertainment, sports, and glamor over intelligence and academics. Being smart or a nerd is looked down upon. What matters is whether you're good-looking and athletic. That's just the way American culture is.
Family reunification is a big issue only for cultures with large extended families. Americans generally have weak kinship ties, and most are opposed to family reunification and the endless chain migration it produces.
It's important to distinguish between reality and the self-serving myths generated by an oligarch-dominated media.
"Engineers have a mean IQ of about 110 (Wolfram, 2023).
The top 0.1% of Americans have IQs of 147 or more. Therefore, the top 0.1% of engineers is ten points to the right, i.e., IQs of 157 or more."
This is wrong on two levels. You assume the two factors, being a 0.1% engineer and being 0.1% IQ are independent, when the reasons IQ might correlate with high performance don't even need to be stated they are so obvious.
Second, you misunderstand how normal distributions work. Simply adding 10 points to the general population's top 0.1% IQ (147) assumes a linear shift at all levels, which isn’t how normal distributions work. This is because the extreme end of the distribution grows narrower and doesn't shift by a fixed amount.
"American college students are now just above the cognitive average, with an estimated mean IQ of 102"
This is debatably true of American college students (there's good reason to believe this estimated mean is incorrect), but is irrelevant for international students. The sorts of people who can manage to achieve high enough grades to succeed in American universities, often not in their native language, sets them a level above the national average (which includes schools all the way down to community colleges). Being an international student has already passed you through a serious qualification procedure.
"Unfortunately, American jurisprudence forbids using IQ tests if the results lead to discrimination on the basis of race, color, religion, sex, or national origin."
This is obviously not true when it comes to accepting foreign immigrants. Our current system very explicitly discriminates based on national origin when we set maximum number of immigrants from any specific country.
1. I'm assuming that American engineer IQ follows the same normal distribution as that of American IQ. If the mean IQ of American engineers is 110, the 0.1% cutoff for them should be ten points farther to the right than the 0.1% cutoff for Americans in general.
If you're having trouble thinking this through, just imagine a picture of a bell curve with its center at 100. Now superimpose the same bell curve on that picture but 10 points farther to the right. The 0.1% cutoff should likewise be 10 points farther to the right.
Now, it's possible that the IQ distribution of American engineers is different. Perhaps it's more skewed. The difference, however, would have to be considerable to change my conclusion. If you still disagree, why don't you crunch the numbers and offer your own estimate? I don't see how it could be substantially different. The total of geniuses (157+) in the entire world cannot be much more than half a million.
2. You're wrong. Domestic students consistently outperform international students.
He, Yunke, and Heather Banham. "International student academic performance: Some statistical evidence and its implications." American Journal of Business Education 2.5 (2009): 89-100. https://eric.ed.gov/?id=EJ1052845
3. In the past, American courts would rule that American principles of non-discrimination do not apply to non-citizens. Recently, there has been a trend to treat non-citizens as if they were citizens.
1. That's a poor assumption. There are simple reasons why a subset of the population would not have a normal distribution. If almost no one of <90 IQ becomes an engineer, this would also boost the average significantly, without changing the IQ of the top 0.1% of engineers by more than a point. What do you think is more likely, that there's a sharp drop in engineers below some certain threshold of IQ, or that the likelihood of someone becoming an engineer is independent of IQ?
But let me run the numbers.
If you preclude all individuals with IQs less than 97, you would get a new, heavily skewed distribution with a mean of 110. While in reality I'm sure there are many programmers with less than 97 IQ, the likelihood of one becoming a programmer probably decreases with decreasing IQ, and my hard cutoff is just a simplification to make calculation easy.
This new distribution (eliminating those with under 97 IQ would create an approximately folded normal distribution) makes it very easy to calculate the SD for a 0.1% occurrence. I get 146.35 from my calculations. Of course in reality it isn't a perfect folded normal distribution, (it would be if the cutoff for a mean 110 IQ was 100), and there's a not-insignificant amount of people in those three points between 97 and 100, so the actual 0.1% would be slightly higher. The real number would be between 148-150, without doing the much more complicated math of calculating SD for a non-standard distribution.
This is all because, while eliminating under performers can significant improve the mean, the upper tail distribution barely changes.
2. Looking more into this, the evidence is extremely limited, with that literally being the only study on the topic of international vs. domestic student performance I can find, which is extremely odd. It seems there may be a taboo on comparing these two groups (not surprising in modern academia), as basically anyone with access to a schools admin software should (in theory) be able to easily compare mean GPA, graduation rate, etc. between the two types of students in an afternoon and make a study about it.
I'll take back my previous statement on this and say I'm unsure. I think there's good reason to be skeptical of this study though, as it seems it only deals with one school, the Okanagan School of Business in British Columbia, Canada, which has a different, and far more easily exploitable international student Visa system than the US.
I'll default to the common knowledge that international students face 1/3-1/2 the acceptance rates of domestic students (no studies I could find estimating this, which may be more evidence of tabboo), especially at top universities, as positions are limited and number of applicants are much higher. More selective generally equates to higher quality, although I can imagine some reasons that this isn't as strong of an effect due to a lower quality applicant pool for international students.
3. This is fair, and potentially a future issue, but discrimination based on national origin is literally the core of the current H1-B system with its national quotas. Any new H1-B program created by the Trump administration will definitely implement national discrimination.
1. The IQ distribution of engineers may be skewed, but even a highly skewed distribution wouldn't change things much. There just aren't that many geniuses in the world, and most are already in affluent countries. It's naïve to think we can "mine" India (or any country) for large numbers of geniuses, and I question the honesty of people who make that argument. They should come clean and say their goal is not immigration of geniuses, but rather immigration of somewhat-above-average people who can do IT grunge work in Silicon Valley.
2. The problem in Canada is not so much that international students are more exploitable but rather that a larger proportion of them don't really see themselves as students. They see the IS program as a means to enter the country and gain permanent status.
3. I honestly don't know. Court rulings are difficult to predict because they hinge on political beliefs, and such beliefs can differ from one judge to another. There will probably be several court decisions before this issue is settled.
1. That’s fair. The 0.1% talent thing is kind of a ridiculous statement, as that’s inevitably going to be 1/1000 of the total number of software engineers, assuming literally all the top software engineers want to come to the US to study, and not stay in India or study in another country like the UK. Having known many international students, they are often obviously very intelligent hardworking people, and any calculation is going to be downwind of that. I still stand by my calculations that the top .1% of engineers won’t have significantly higher IQ than the top .1% of the population in general though, based off my assumptions.
2. I meant more easily exploitable by the students. You’re definitely right that most don’t treat it as coming to study but just coming to immigrate. I believe there’s no cap on the number of years you can take to finish a degree in their system (or its high like 8 years) so students take 2 credits a semester or something while working full time.
I think It'd be more than 7,000,000 because (10^9)*(100-99.993) = 7 million. 157+ is a bit less than 4 standard deviations in China, and China has like 1 billion people (order-of-magnitude estimate purposes). That being said, I get the (extremely uninformed, mind you) impression that there is a lot of focus is on getting talent from India, ignoring the fact that most of the cognitive elite of that country might already be overseas.
Moreover, the U.S. definitely lacks the ability to attract most of the cognitive elite in China.
Also, I wonder to what extent the historically high levels of endogamous stratification in many regions of the Third World, such as Nigeria or Saudi Arabia, affect the existence of a cognitive elite in those countries. Perhaps these regions have thicker tails (from both high and low castes) than one would expect from a normal distribution.
Nvm, I'm clearly innumerate. Four standard deviations is way more extreme than that. I should have checked with R instead of trusting information online...
I will let the author take this on and limit myself to saying that you are missing the main point.