Branko Milanovic has a piece in the FT (gated) where he advocates a form of differentiated citizenship as a way to make immigration more acceptable in rich countries. Milanovic argues that immigration is one of the most powerful mechanisms to reduce global income inequalities (by allowing inhabitants of poor countries to increase their income by moving to rich countries) but it is susceptible to create tensions within inhabitants of rich countries. Hence, we should find ways to make it more acceptable within rich countries, and this should be done by essentially restricting the rights of migrants so that citizens do not need to share the “premiums” reserved to them.

Chris Bertram has a very critical response to the piece, claiming that Milanovic advocates a form of apartheid, and that there are a number of inalienable rights that cannot be violated even for purposes of economic outcomes. Moreover, the differentiation in rights proposed by Milanovic has already been in place in a number of countries. Germany for instance has a huge population of “denizens”, people of the 2nd or 3rd generation (children of guest workers that came in the 1960s and 1970s) that do not enjoy citizenship.

My take on this is that Milanovic is is wrong to think that differentiating rights will eliminate tensions and reduce opposition to immigration, but many advocates of equal rights for migrants also fail to see that equality would mean less immigration (and less global inequality reduction) in the first place. What I’m arguing here is based on a conference paper that I just presented in Philadelphia. It seeks to explain why some countries (Switzerland, Germany) massively resorted to foreign labour to fill in labour needs, while others (Sweden, but most of Scandinavia as well) did not.

First, differentiating rights makes migrant workers much cheaper than native workers, and is therefore likely to exacerbate tensions rather than dampen them. Germany and Switzerland had guest-worker programs where immigrants not only had no limited or no right to family reunification, but were also not allowed to change employer because their immigration status was tied to a specific job. If you cannot leave your job to get a better one, your bargaining position is weaker than other (native workers), and you will be more attractive for employers than natives. Hence, differentiating rights is likely to create a form of unfair competition that may actually exacerbate xenophobic tensions. You can hear this regularly in the United Kingdom, where UKIP-leaning working class natives complain about immigrants accepting lower wages. Differentiating rights exacerbates this.

Second, one has to accept that conferring equal rights to migrants would probably mean lower immigration levels in rich countries, and therefore less international income redistribution. Milanovic gives the example of the Gulf States, that give very few rights to migrants while having a huge proportion of them. There is clearly a connection between these two factors: migrants are so attractive as a source of labour because they have lower reservation wages and can be dismissed easily during economic downturns. Hence, the social cost of their employment doesn’t need to be assumed by the host country. It is fairly easy to imagine that the Gulf States would import much less foreign labour from the Philippines, Pakistan or elsewhere if they had to pay for healthcare for their children, and fund a system of unemployment benefits in case their jobs disappeared. Because labour would be more expensive, they would probably seek to mechanize more, use fewer workers and less income would be sent to poor countries in the form of remittances.