Last night I went to Bristol Freedom Society to debate Ryan Bourne, of the Centre for Policy Studies, over whether the UK should have open borders. It was very enjoyable, and while I won very marginally, convincing one person to switch sides from closed to open borders, I thought it was extremely close, and Ryan was certainly a very fluent and convincing speaker.
My main case was (a) restricting migration restricts extremely important rights, like the freedom to take a job you are offered and the freedom to offer a job to a desired applicant, (b) when we curtail these sorts of freedoms we need to have a preponderance of evidence that the costs are very high, (c) the economic evidence says immigration is pretty good for the recipient country, very good for the source country, and amazingly good for the migrant themselves, (d) the magnitude of the social/cultural impact (i.e. the effect of migrants on our institutions, customs, etc.) is unclear, (e) therefore, we ought to have open borders (or something very close to open borders).
Ryan's counterargument centred on the claim that the benefits of restricted migration would not extend if it was unrestricted or close to unrestricted, because migration of certain amounts undermines the institutions that cause migration to benefit people at all. (It was more complex, but this formed the nub of the debate). As is suggested by my point (d), I think there is some plausibility to this argument, but I also think it is under-studied. I suggested this when we were able to discuss the points we'd made in our opening statements a bit more, and we had a lot of back-and-forth over the issue, but we didn't resolve our disagreement. Without trying to guess at Ryan's position or put words in his mouth, I will stake out three claims I think we must accept to have the debate within a rational framework.
1. All positions are on a continuum from complete open borders (as much gross immigration as non-natives wish/can afford) to complete closed borders (no gross immigration). Perhaps the best way of accurately describing positions is by how much migration they favour. Open borders advocates favour something close to 100% of the amount that would occur under open borders. Everyone else is somewhere on a spectrum from no gross migration to the 100% open borders case (past societies have also had forced immigration, i.e. slavery, so these aren't the only theoretical positions, just the only persuasive ones.)
2. The supposed socio-cultural problems of migration come from particular numbers of migrants. No one thinks 5,000 migrants a year to a country the size of the UK will fundamentally undermine its customs, laws, institutions and so on. Many people think 5,000,000 migrants per year would. So to be anti-open borders you implicitly have to have an estimate of how much migration you think is going to occur. If open borders only led to 5,000 migrants per year, then almost no one would be against it. It is because open borders would be expected to lead to too much migration that people oppose it. This doesn't change if it's a question of probability distribution—then migration is opposed because it raises the chances of too much migration occuring to too high a level. Everyone must (at least implicitly) have an expected level of migration to oppose open borders.
3. Any claim that migration should be kept to a particular level, because of the risk of undermining British institutions, implies an assumption about how much damage the marginal immigrant does or will do (reliably or with some probability). One cannot cop out of the question, you need to have an answer. But no one has yet set out good evidence about exactly how much damage to institutions the marginal immigrant does or will do—typically arguments in this area depend on anecdote or things that people feel they "just know". This won't do when the benefits to immigration are so high. We cannot simply assume the cost to our institutions outweighs the other benefits.
I think once these three points have been accepted, there is a lot of room for good empirical work. But until they have, a lot of the migration debate will be unclear, vague, and people will be talking past one another.