It’s been a while since I blogged here and I figure that some shameless self-promotion is the best way to ease back in. As discussed in Jason’s review, I’ve got a new book out titled Unequivocal Justice. The book builds on work from other BHL contributors, including Jason, Jacob Levy, and Will Wilkinson.
“Ideal theoretical” analyses of political institutions face a dilemma. In an ideal world in which everyone fully complies with the principles of justice, then coercive state intervention isn’t needed to secure justice. People will do the right thing out of the goodness of their hearts. On the other hand, in a nonideal world where people are not fully compliant with justice, state intervention may have a role to play, but the people running the state itself might not act justly. In brief: if we rule injustice out, then the state isn’t needed; if we rule injustice in, then we may not assume that the state itself is just.
This dilemma poses a problem for an ideal theorist like Rawls who assumes that the state is both needed to secure justice and will operate justly and effectively. Rawls and others resolve the dilemma illicitly; namely, they rule injustice in to give the state a job to do and then (implicitly) rule injustice out to ensure that the state does that job well.
The public goods argument for the state makes the problem particularly clear. Rawls starts with a textbook explanation for why people won’t voluntarily contribute to public goods—they have a strong incentive to free ride. As Rawls puts it, “Where the public is large and includes many individuals, there is a temptation for each person to try to avoid doing his share. This is because whatever one man does his action will not significantly affect the amount produced” (A Theory of Justice, 236). Better to let everyone else pony up for a Tesla and buy a cheap gas guzzler for yourself; this way you get the clean air produced by others without paying the costs yourself. But everyone else is thinking the same thing, so they all buy gas guzzlers, too. The result is polluted air for everyone. Since the market won’t provide public goods, Rawls concludes that “the provision of public goods must be arranged for through the political process and not through the market” (A Theory of Justice, 236). For instance, the state can enforce a cap-and-trade system to reduce emissions.
How, though, will the political process arrange for the provision of public goods? Suppose that the election pits a candidate who favors cap-and-trade against a candidate who favors no environmental protection whatsoever. Will I, as a voter, do my share and contribute a vote for the cap-and-trade candidate? By Rawls’s own assumptions, the answer is “no.” Better to let everyone else research the candidates’ platforms, read the experts’ opinions on the effectiveness of cap-and-trade, and spend time at the polls while I stay at home watching TV (or perhaps cast a thoughtless, uninformed vote at the last minute). But everyone else is thinking the same thing, and so they too fail to vote for the cap-and-trade candidate. Thus, the behavioral assumption that Rawls invokes to generate a need for state intervention—that people are motivated to free ride—simultaneously undermines the state intervention itself. People have just as much temptation to free ride on the thoughtful, informed votes of others as they have to free ride on the emission reductions of others.
To resolve this problem, Rawls (implicitly) resorts to using different behavioral assumptions to model nonpolitical behavior and political behavior. We free ride in the market but not in politics. But this move violates what Geoffrey Brennan and James Buchanan call “behavioral symmetry”—that is, we should apply the same behavioral models across different institutional contexts to ensure a fair, apples-to-apples comparison of those institutions.
Here’s how Rawls will reply. Ideally just people are only “conditional cooperators.” They prefer to do their share so long as others do the same. But they won’t be “suckers” and contribute when others aren’t contributing. Thus, they prefer to reduce their emissions when others also reduce their emissions, but they need state enforcement to get assurance of reciprocation.
This reply won’t work because the state introduced to solve the assurance problem simply creates another assurance problem. Think back to the voting case. Suppose you’re on your couch deciding whether to spend the day with your family at a park or to spend it looking up experts’ opinions about environmental policy, candidates’ voting records on the environment, and so on. You are happy to do the latter so that you can cast a thoughtful, informed vote on election day—but only if you have assurance that others will too. However, no such assurance is available, and so you head off to the park instead of putting time and effort into your vote. Other citizens do the same and, as a result, just and effective environmental policy goes unprovided. Here again, Rawls’s reason for thinking that state intervention is needed is equally a reason for thinking that the intervention won’t work.