Reconciled

In Political Liberalism, John Rawls revises his theory of justice by adopting a stronger position on the priority of political liberty, which includes not merely the right to vote but the right to participate in the electoral process, to run for office, and so on. He writes,

… we must take an important further step and treat the equal political liberties in a special way. This is done by including in the first principle of justice the guarantee that the political liberties, and only these liberties (emphasis mine), are secured by what I have called their ‘fair value.’ (Rawls 2005, 327)

Rawls’s argument, in his own words, is that “unless the fair value of these liberties is approximately preserved, just background institutions are unlikely to be either established or maintained.” To figure out how to pull this off, we can’t say much in advance, but “must advance by trial and error.”

Rawls does suggest that “one guideline for guaranteeing fair value seems to be to keep political parties independent of large concentrations of private economic and social power …” (328). But he declines to say more.

Rawls rejects some reasons for prioritizing political liberty, such as the claim that it is the best way to secure autonomy, since he recognizes some reasonable people will deny this. Instead, political liberties are “essential in order to establish just legislation and also to make sure that the fair political process specified by the constitution is open to everyone on a basis of rough equality.” (330)

I want to raise a difficulty for Rawls’s argument. Jason Brennan has recently argued, I think correctly, that Rawls’s arguments for specific rights don’t work, in part because they treat liberties asymmetrically in ways that cannot be justified. Let’s suppose Rawls can answer this generic challenge and see if he can avoid a more targeted objection.

Here’s the objection: all of Rawls’s arguments for a right to influence government applies not merely to voice mechanisms – voting and the like – but to exit mechanisms as well, such as a right to emigrate and free movement, rights to legal exemptions, rights to federal arrangements, and rights of secession. Ilya Somin aptly calls rights of freedom of movement “foot voting” because people are exercising liberty to shape political institutions by withdrawing from them, rather than participating in them. Both voice and exit mechanisms can greatly influence government behavior when people pursue them in large numbers.

Both types of liberty can be “essential in order to establish just legislation” and to ensure that the political process is one of “rough equality.” In many cases, the threat of exit pressures governments to improve more than the threat of losing votes. And if people were freer to move between countries, that pressure would arguably increase.

So Rawls’s arguments seem to ground both democratic rights and exit rights. If so, the institutional implications of Rawls’s arguments in Political Liberalism change, even if we stick to ideal theory, and even if we restrict the range of reasonable pluralism to conceptions of the good (once you relax ideal theory assumptions somewhat, and extend reasonable pluralism more broadly, I think the case for exit rights grows even stronger). It will be important to decentralize government, and ease freedom of movement, perhaps even with subsidies.

Here are some potential replies.

1. Exit and Voice Mechanisms Compete: as Hirschman pointed out long ago, the more people exit an institution, the less effective voice mechanisms become, at least some of the time. If people can leave, why should they invest in improving the institution rather than leave it? So perhaps we need to restrict exit rights in order to improve voice rights. In reply, I’d point out you can make a parallel argument in the other direction: if you restrict exit, you make exit less effective, so this isn’t a reason to prioritize voice over exit, and certainly not to prioritize it as much as Rawls does.
2. Voice is More Effective than Exit: one might argue that democratic voting is a more effective means of communicating dissatisfaction and the general will of the people than exit mechanisms. I think Somin has shown this just isn’t so by drawing standard rational choice theory to explain political ignorance (following a long line of political scientists). Democratic voting isn’t as responsive to legislative changes as one might hope. Exit mechanisms, on the other hand, often convey more information (since it is odd to expressively exit rather than expressively vote) and deprive governments of more power.
3. Exit Mechanisms are More Biased in Favor of the Rich Against the Poor:  it may seem that the rich have a relatively easier time exiting than the poor vis-a-vis voting than the poor. But Rawls knows that the rich can dominate democracy too, and so ideal regimes will restrict the influence of the rich. Why not think that ideal regimes can similarly restrict the influence of richer exiters? The government could also equalize opportunities for the poor and rich to exit with redistributive taxation, say granting the poor subsidies to move across the country, or to push for decentralization, or even to leave the country.

There are other potential replies, but these are enough for now to get my general point across.