I promise that the ellision of paragraphs of text did not cause the weakness of the argument below.
When a person increases his earnings by moving to the United States, or to Sweden, or to China for that matter he or she does so by creating more value than was possible in the previous context.
a policy regime that’s generous to wannabe migrants makes everyone much better off than a policy regime that keeps people trapped.
So if world GDP increases then ... then everyone is better off.
I am shocked to report that I am quoting Matthew Yglesias ?!?
I commented on his blog.
Freudian Slip ? When I read "The general point that needs to be kept in mind is that the *global* economic space is not zero-sum. " I thought you were making a point which you ignore.
If the US welcomes highly trained immigrants then less educated people in extremely poor countries, which invested scarce resourses in their training, will suffer. Your "global" perspective considers various people in the USA. It does not specifically consider the less skilled in the countries of origin of skilled migrants.
You do mention the world, but only when discussing world money-metric utility. The dollar value of work in the new country is greater than the dollar value of work in the old country so ... "everyone is better off."
I'd consider that underpants gnome welfare economics.
In no other context do you display complete total indifference to the distribution of income. In this post, you assert that if world GDP is increased, then everyone is better off. That is very interesting meta-ethical nonsense.
I see a real dilemma. People should be free to leave a country even if they are valuable to that country -- that is a core principle of liberalism -- that people belong to themselves and are not slaves of the collectivity. On the other hand, this can be very bad for utilitarian welfare -- the world sum of utils -- if it means a brain drain from human capital poor countries. I think this is one of many cases in which it is hard to be both a consequentialist and a liberal. I think the cognitive dissonance was enough to make you temporarily forget that world income is not world welfare and that income distribution matters -- no that's impossible -- to make you temporarily seek comfort from arguments which are based on that hidden assumption.
Yglesias seems to be using the usual trick of noting that a policy regime with free migration and lump sum transfers to those who lose from the migration and lump sum taxes on those who gain can make everyone better off. So it isn't that distribution doesn't matter, but that we can make the distribution whatever we want without deadweight losses, because, without making taxes conditional on any data (they are lump sum) we can set them just exactly right.
This idea is totally absurd. The implicit proposal -- that taxes be charged to people by name with no need to justify why one person pays more than another, is clearly unconstitutional (a bill of attainder). It is one of the tricks used by economists who don't give a damn about anyone but the very rich to con the masses.
It leeds immediately to the idea that the income tax rate should be zero as should the consumption tax rate. Yglesias knows it's a trick.
addendum more stuff not in the comment
Yglesias's conclusion was sloppily phrased. It should have been "for any policy regiem that keeps people trapped, there is a policy regime that’s generous to wannabe migrants makes everyone much better off." The claim as written is plainly false, since it means that any policy regime with free immigration (including the one which goes on to launch nuclear weapons on all the worlds largest cities) is better than any policy regime without. The rewritten claim is standard in the economics literature but it depends on a definition of "policy regime" which is not standard or consistent with the US constitution or basic principles of the rule of law. The regimes in question typically require absolute arbitrary power to take anything from anyone.