Before you start reading, this post may irritate you if you’re a libertarian. so take this as a warning. But, as a libertarian sympathizer,1 all I’m really doing is playing devil’s advocate. Rather than me making a point, interpret this post as a set of challenges (questions) that invite your answers.
One of the most ambiguous terms in the libertarian lingo is “limited government.” What’s meant by the word “limited?” If it means constrained, well, all governments are constrained to one extent or another. Of course, some governments are more constrained than others, and in different ways. When a libertarian uses the term “limited government” she probably has in mind governance more limited than the modern-day welfare state democracy. Still, what exactly does this mean? What should we limit, and by what rule and/or process will society decide where to limit and where to extend the power of government? As a liberal, I contend that the limits on government are emergent phenomena, in part decided through reasonable public debate. I think the implications of this may upset certain libertarian designs.
Let’s approach the problem from a basic public choice angle, as introduced in Buchanan and Tullock’s The Calculus of Consent. The idea in Buchanan and Tullock (1962) is that a group of individuals can agree to a constitution that dictates the rules and process of political decision-making. On this basis, they divide the problem of collection action into two stages: the constitutional level and the operational level. The latter is the day-to-day decision-making a government will make. Allowing for disagreement on policy at the operational level, by agreeing to a set of rules (a constitution) society then use a process to settle political disputes. Since political preferences are heterogeneous, disputes at the operational level can be resolved through trade, such as vote trading (if you vote for A, I will vote for B). This, it should be clear, is an abstract theory that describes an ideal.2
The explicit assumptions I’m making are,
- If there are things that can be solved through collective action with relative efficiency, a group of people have an incentive to form a government;
- Knowing that there will be disagreement over specific policies, this group of people can unanimously (or near unanimously) decide on a constitution, which defines a set of rules that decide decision-making at the operational level;
- In exchange for policies you support, you will accept the external costs of policies you disapprove of.
There will be policies legislated that you disagree with and they come at a cost to you. But, governments can be organized in a way such that only a small minority of people actually participate in the decision-making: representative governments. For example, (this is from Buchanan and Tullock ) a bicameral legislature, where the voter sets for members of one house will be different from their respective peers of the other, will be closer to unanimity than a government without one, all else equal. As the institutions of governance become more complex, the average person loses track of the individual exchanges, so that you may not realize that the burden you suffer is in exchange for some benefit you enjoy. Instead, you may see only the injury. Suppose that we know our current institutions are imperfect. As scientists, how accurately can we judge between perfectly inclusive policies (those that would exist under unanimity) and partially (or completely) externalized policies? Rephrased, if what we’re interested in is a pluralist society, how can we claim with any certainty that, for example, “wealth transfers to the poor” is an imperfectly inclusive policy?
Typically, when someone defines “limited government” they restrict the scope of the state to national defense, currency control (i.e. if you think money is a public good), and/or justice. The exact definition may vary on what set of services are categorized as public goods. But, that’s the thing. There’s reasonable disagreement over what service is a public good, and there’s reasonable disagreement over how to best solve the public good (including the option that the cost of collective action is too high). These disputes have to be settled through a pluralist decision-making process, and the ultimate product will differ by some degree from your ideal. The limited (perfectly pluralistic) government of the real world will provide goods and services which are, at least partially, different from the set you would prefer. It may be the case that the real world limited government is larger than your ideal (a larger set of public goods is provided for).
In summary, I have two challenges for fellow libertarians (or liberals with libertarian preferences).
Challenge #1: When judging collective decisions, do libertarians suffer a fatal conceit? The more complex political exchange is, the less the average person understands the system as a whole, and therefore the less able we are to know which, or how much of a, policy is non-inclusive.
Challenge #2: In a perfectly pluralistic society, is it the case that a public with reasonable disagreements may ultimately decide on a government that differs from each individuals’ own ideal?
What I’m interested in, though, is if libertarians should start thinking about that margin of error in their judgment. For example, how unreasonable would it be to believe that progressive redistribution would exist in a perfectly plural society? I can think of a strong public goods argument for progressive redistribution of wealth. If in the day-to-day decision-making we come to compromise over disagreements, we may end up adopting policies that we disagree with for the sake of pushing legislature we agree with. Doesn’t this mean that the freest (voluntary governance) society could very well have a welfare system? Or, be responsible for certain infrastructure investments? Now, what the institutions of this government would look like is a completely different question. Maybe we’ll see greater competition between governments, if, for instance, territorial boundaries begin to lose relevance (greater cosmopolitanism). I don’t think this last question is answerable anyways.
1. I made the term “libertarian sympathizer” up. I consider myself, first and foremost, a liberal, in that I support governance that approaches unanimity as much as possible. What unanimity means is that everyone’s opinion counts in collective action. But, I support a set of policy prescriptions that is broadly libertarian — free banking, limiting the size and scope of the military, ending corporate welfare, et cetera. I just don’t think these policies are the ones that should be implemented over all other policies: these things should be up for public debate, the only liberal position.
2. As Buchanan and Tullock wrote,
The relevance of the contract theory must lie, however, not in its explanation of the origin of government, but in its potential aid in perfecting existing institutions of government.
— James M. Buchanan and Gordon Tullock, The Calculus of Consent (Indianapolis: Liberty Fund, 1999 ), p. 319.