As for why immigrants (or their employers) should buy insurance when natives don’t have to, note that immigrant insurance covers only risks added by migration! People (foreigners or citizens) who remain in their native countries instead of migrating would not have to buy immigration insurance because they do not add any risks in the destination country (the base costs of the risks they pose in their own countries are covered by local arrangements). Consider a hot-tempered Pakistani who poses a 1% annual risk of murdering someone. If he remains in Pakistan the annual expected value of his bad temper to his neighbors is 0.01 times less than $1 million, but when he enters the USA it becomes 0.01 times more than $9 million. Furthermore, in Pakistan, however much harm that hothead inflicts on his neighbors, he can inflict virtually no harm on Americans. Asking that Pakistani to carry Sudden Jihad Syndrome insurance while he is in America is like asking him to carry liability insurance while he drives a car in America. The insurance is for incremental risk. (Note that immigration insurance might as well be symmetrical– if the USA requires it other countries may also.)
The fact that different immigrants, individually or (for example) by national groups pose different risks would be reflected in their insurance premiums, more or less accurately depending on actuarial considerations and how much regulatory “anti-discrimination” foolishness was injected into premium-setting. High premiums might discourage some prospective migrants. That would be a good thing– forcing immigrants (or their employers) to internalize costs which they now palm off on everyone else. Of course, if insurers were required to “community rate” all immigrants equally instead of distinguishing Brahmins from Salafis, the system would not work as well. If immigrants really aren’t very dangerous then their insurance premiums won’t be very high.
In fact, the big cost would not be insuring against criminal behavior, but against social spending (welfare) consumption. “Community-rated” premiums for that would track the mean expected outlay, which is around $14,000 annually for households in the income range of the mean immigrant household of about $60K (citations on request). Individually-underwritten premiums would vary a lot– immigrant scientists, for example, might pay very little. If we truly cut off social spending on immigrants, we wouldn’t have to ask for insurance against it. (Despite the oft-repeated big lie that immigrants are “not eligible for welfare,” they actually are. Really. Please don’t make me provide the citations again.)
Sailer didn’t offer a detailed proposal. He was trying to put across the point that some immigrants inflict outsized costs– way more than their individual labor is worth– and those are currently absorbed by citizens at large instead of by the folks who benefit from immigration (the immigrants themselves and employers). Employers want to privatize profits (cheap labor) and socialize costs (e.g., Obamacare subsidies for immigrants [official link]). Textbook theory says we should press them to internalize all their costs. There is an insurable risk, and the premiums should be paid by immigrants. For those who work the costs would pass through to their employers and for those who don’t (currently 2/5) the costs would fall on whoever supports them. Employers certainly could pay premiums for SJS insurance. The moral hazard would be small because employers do approximately nothing to control immigrants now so they could hardly do less.
