This is a philosophical case for open borders that builds on the philosophy of John Rawls, particularly the veil of ignorance argument, which Rawls borrowed from John Harsanyi. According to Wikipedia:

It is a method of determining the morality of a certain issue (e.g. slavery) based upon the following thought experiment: parties to the original position know nothing about their particular abilities, tastes, and position within the social order of society. The veil of ignorance blocks off this knowledge, such that one does not know what burdens and benefits of social cooperation might fall to him/her once the veil is lifted. With this knowledge blocked, parties to the original position must decide on principles for the distribution of rights, positions and resources in their society. As Rawls put it, “…no one knows his place in society, his class position or social status; nor does he know his fortune in the distribution of natural assets and abilities, his intelligence and strength, and the like.”[5] The idea then, is to render moot those personal considerations that are morally irrelevant to the justice or injustice of principles meant to allocate the benefits of social cooperation.

One of the problems with making a Rawlsian case for open borders is that Rawls himself was interested in confining his moral analysis within nation-states. This issue was brought up in the blog post Against Social Justice by Jacob Levy. However, there is no intrinsic problem with applying the Rawlsian veil of ignorance approach at a universalist, rather than nationalist, level, even if Rawls himself didn’t do this.

In an article titled Aliens and Citizens: The Case for Open Borders, Joseph Carens attempts to appropriate the Rawlsian framework to determine the morality of open borders. A relevant portion of Carens’ argument is quoted below:

In contrast to Nozick, John Rawls provides a justification for an activist state with positive responsibilities for social welfare. Even so, the approach to immigration suggested by A Theory of Justice leaves little room for restrictions in principle. I say “suggested” because Rawls himself explicitly assumes a closed system in which questions about immigration could not arise. I will argue, however, that Rawls’s approach is applicable to a broader context than the one he considers. In what follows I assume a general familiarity with Rawls’s theory, briefly recalling the main points and then focusing on those issues that are relevant to my inquiry.

Rawls asks what principles people would choose to govern society if they had to choose from behind a “veil of ignorance,” knowing nothing about their own personal situations (class, race, sex, natural talents, religious beliefs, individual goals and values, and so on). He argues that people in this original position would choose two principles. The first principle would guarantee equal liberty to all. The second would permit social and economic inequalities so long as they were to the advantage of the least well off (the difference principle) and attached to positions open to all under fair conditions of equal opportunity. People in the original position would give priority to the first principle, forbidding a reduction of basic liberties for the sake of economic gains.[5]

Rawls also draws a distinction between ideal and nonideal theory. In ideal theory one assumes that, even after the “veil of ignorance” is lifted, people will accept and generally abide by the principles chosen in the original position and that there are no historical obstacles to the realization of just institutions. In nonideal theory, one takes account of both historical obstacles and the unjust actions of others. Nonideal theory is thus more immediately relevant to practical problems, but ideal theory is more fundamental, establishing the ultimate goal of social reform and a basis for judging the relative importance of departures from the ideal (e.g., the priority of liberty).[6]

Like a number of other commentators, I want to claim that many of the reasons that make the original position useful in thinking about questions of justice within a given society also make it useful for thinking about justice across different societies.[7] Cases like migration and trade, where people interact across governmental boundaries, raise questions about whether the background conditions of the interactions are fair. Moreover, anyone who wants to be moral will feel obliged to justify the use of force against other human beings, whether they are members of the same society or not. In thinking about these matters we don’t want to be biased by self-interested or partisan considerations, and we don’t want existing injustices (if any) to warp our reflections. Moreover, we can take it as a basic presupposition that we should treat all human beings, not just members of our own society, as free and equal moral persons.[8]

The original position offers a strategy of moral reasoning that helps to address these concerns. The purpose of the “veil of ignorance” is “to nullify the effects of specific contingencies which put men at odds” because natural and social contingencies are “arbitrary from a moral point of view” and therefore are factors which ought not to influence the choice of principles of justice.[9] Whether one is a citizen of a rich nation or a poor one, whether one is already a citizen of a particular state or an alien who wishes to become a citizen—this is the sort of specific contingency that could set people at odds. A fair procedure for choosing principles of justice must therefore exclude knowledge of these circumstances, just as it excludes knowledge of one’s race or sex or social class. We should therefore take a global, not a national, view of the original position.

Those in the original position would be prevented by the “veil of ignorance” from knowing their place of birth or whether they were members of one particular society rather than another. They would presumably choose the same two principles of justice. (I will simply assume that Rawls’s argument for the two principles is correct, though the point is disputed.) These principles would apply globally, and the next task would be to design institutions to implement the principles—still from the perspective of the original position. Would these institutions include sovereign states as they currently exist? In ideal theory, where we can assume away historical obstacles and the dangers of injustice, some of the reasons for defending the integrity of existing states disappear. But ideal theory does not require the elimination of all linguistic, cultural, and historical differences. Let us assume that a general case for decentralization of power to respect these sorts of factors would justify the existence of autonomous political communities comparable to modern states.[10] That does not mean that all the existing features of state sovereignty would be justified. State sovereignty would be (morally) constrained by the principles of justice. For example, no state could restrict religious freedom and inequalities among states would be restricted by an international difference principle.

What about freedom of movement among states? Would it be regarded as a basic liberty in a global system of equal liberties, or would states have the right to limit entry and exit? Even in an ideal world people might have powerful reasons to want to migrate from one state to another. Economic opportunities for particular individuals might vary greatly from one state to another even if economic inequalities among states were reduced by an international difference principle. One might fall in love with a citizen from another land, one might belong to a religion which has few followers in one’s native land and many in another, one might seek cultural opportunities that are only available in another society. More generally, one has only to ask whether the right to migrate freely within a given society is an important liberty. The same sorts of considerations make migration across state boundaries important.[11]

Behind the “veil of ignorance,” in considering possible restrictions on freedom, one adopts the perspective of the one who would be most disadvantaged by the restrictions, in this case the perspective of the alien who wants to immigrate. In the original position, then, one would insist that the right to migrate be included in the system of basic liberties for the same reasons that one would insist that the right to religious freedom be included: it might prove essential to one’s plan of life. Once the “veil of ignorance” is lifted, of course, one might not make use of the right, but that is true of other rights and liberties as well. So, the basic agreement among those in the original position would be to permit no restrictions on migration (whether emigration or immigration).