Last month, I presented a paper at a Georgetown University conference on “The Ethical Limits of Markets.” The title of my paper was “A Libertarian Case for the Moral Limits of Markets,” but really the paper is a call for libertarians to think about markets less in terms of what we philosophers call “ideal theory” and more in terms of “non-ideal theory.”

Like most proponents of radical political ideologies, libertarians are exceedingly good at describing their vision of an ideally just society. Unfortunately, libertarians also have in common with other radical visionaries a tendency to be a little fuzzy on the details about how we get from here to there.

I’m not talking about strategy. Libertarians like to talk about strategy – about whether its better to work within the system or outside it, to build alliances with the left or the right, whether to look for agreement with people on single issues or try to sell them on the fundamental moral and economic principles, and so on.

No, what I had in mind is a different sort of question. Suppose we all agree that libertarianism is right about what an ideally just society would look like. What does that imply about what justice looks like now, in our decidedly non-ideal reality?

This question, I think, is harder than it looks. It might seem obvious, for instance, that libertarians should be in favor of any policy change that moves us toward their ideal, and opposed to any change that moves away from it. Libertarians like free markets. Real world markets aren’t free. But any policy move in the direction of freer markets has to be something libertarians should applaud, right?

I think there are a few problems with this idea. First, it’s not always obvious (or maybe even determinate) whether a policy change ought to count as a step in the direction of freedom or not. Consider the privatization of US prisons. Should libertarians celebrate the fact that many US prisons are now run by private companies rather than by the state itself? Nominally, such privatization involves taking some power out of the hands of the state and into the hands of market actors. And because prison contracts are assigned on the basis of a competitive bidding process, privatization also arguably keeps costs lower than they would be under a system of state monopoly.

But as both Charles Johnson and Sheldon Richman have noted, there are serious problems with this superficial analysis. First, prison privatization, like much so-called “privatization” in the United States, doesn’t really eliminate the state’s monopoly on prison services. It merely transfers that monopoly into nominally private hands, leaving the state with the power to dictate many of the significant details of how the business is to be conducted.

Second, a lot of what prisons do, whether they’re public or private, is unjust from a libertarian perspective. Many prisoners in the United States today are non-violent drug offenders. Those people should not be in jail at all – holding them there is, from a libertarian perspective, a violation of their rights. The fact that under a regime of privatization their rights are violated by market actors rather than the state shouldn’t make that regime any more attractive from a libertarian perspective.

In fact, and this is my third point, it arguably makes it worse. If privatization really does keep the cost of running prisons down, then privatization allows the government to engage in more of an unjust activity than it otherwise might be able to do. If the government is violating people’s rights, do we really want to help them be more efficient about it?

So if the goal is freedom, it’s sometimes hard to tell whether a particular policy change – privatization, vouchers, deregulation, etc. – is a step in the right direction or not. But there’s another problem, too.

Suppose you have a policy change that is unequivocally a step in the direction of greater freedom. Suppose that a particular state prohibits all of its citizens from engaging in any economic activity on Mondays. Then legislation is proposed that would allow white men, but only white men, the liberty of doing business on any day they choose. Is this a policy change that libertarians ought to support?

In terms of freedom, it’s a Pareto-superior move. Some people’s freedom is increased, and nobody else’s is diminished. Still, however, the policy change offends against an idea that, I believe, is an important part of the libertarian intellectual framework, even if we libertarians sometimes marginalize or lose sight of it. That idea is equality before the law. Libertarians don’t merely believe in freedom; they believe in equal freedom. That’s part of the reason why libertarians should be in favor of marriage equality, even if they don’t think the state has any business being involved in marriage in the first place. (On these points, see Steve Horwitz’s excellent pieces here and here). Policies that increase freedom can nevertheless run afoul of the ideal of equality before the law, and thus, from a libertarian perspective, be undesirable.

Libertarians disagree with each other about the exact shape of their ideal. Some want a minimal state, some want a somewhat larger state, and some want no state at all. And we spend a lot of time arguing with each other about those differences. But even if all of those arguments were to be resolved, there would still remain important philosophical questions about how to get from here to there. What’s the just way to act in a deeply unjust society? What does justice require in an imperfect situation, when the full demands of justice simply aren’t a viable option?

More in the paper.