How to Argue for Libertarianism

or anything else

There are two ways to defend any political position: Moral arguments or economic, more broadly consequentialist, arguments. The moral argument for libertarianism usually starts with the idea of negative rights, rights not to have things done to you. Moral arguments for other political positions sometimes start with positive rights, rights to get something, enough food, good medical care, an education. Other positions can be defended by claims of obligation to your sovereign, your country, your people.

Moral Arguments

Moral claims are rhetorically effective when preaching to your fellow believers but not very useful for convincing unbelievers since we have not yet come up with any way of showing what moral claims are true, despite several thousand years spent trying; moral philosophy is not one of the more rapidly progressing fields. Philosophers still read Aristotle, physicists and economists do not.

A second problem is that the more carefully you look at the moral claims of any political position, the less clear they look. I know that best for my own position; I spent one past post on libertarian problems. They include the problem of initial appropriation (how, morally speaking, did I get ownership of my land), what are the limits of my claim against someone who violated my rights, how large does a trespass have to be to count (trespassing photons), what if a small violation of someone’s rights produces enormous benefits, human shield problems … . It is possible that a sufficiently able libertarian philosopher could come up with a version of libertarian moral philosophy that dealt adequately with all of these, but I don’t know that any has and, even if it can be done, the random libertarian hasn’t done it. I will leave to progressives, Marxists, traditionalist conservatives, monarchists and nationalists the project of pointing out similar problems with the moral claims that support their political positions.

Consequentialist Arguments

The alternative to a moral argument is a consequentialist argument, an argument offering reasons to believe that your preferred political system will produce better results than alternative systems. Since I am not only an economist but an economic imperialist, believe that economics is useful for understanding practically anything that depends on human behavior— my first journal article in the field was an economic theory of the size and shape of nations — and some things that don’t, I mostly think of arguments about consequences as economic arguments.

One problem with the consequentialist approach is that “better” in “better results” is a moral term. Without moral arguments to identify good and bad how can I know what results are better, what worse? The answer is that I can leverage the existing moral beliefs of the people I am trying to persuade. I don’t have to show that the outcomes of libertarian policies are good in the mind of God, only that they are good in their eyes. People do not all have the same moral beliefs but at the level of judging outcomes there is a lot of overlap.

Very few people approve of hunger, poverty, sickness, misery. If alternative political systems produce only slightly different outcomes, one a little less hunger at the cost of a little less freedom, a libertarian and a utilitarian might disagree about which was better. But if, as I believe, more freedom usually results in less hunger, less ignorance and less misery, the two can agree on preferring it without agreeing on a common set of values. If, as I suspect, most people value most of the same things, even with different weights, and if my preferred political and legal institutions produce results that are not merely a little better than the results produced by alternative institutions but much better, it is likely that most people convinced of my factual claims would agree with my political conclusions. I only have to persuade them of my economic arguments, can rely on their existing values.

I have been arguing politics, at this point, for some sixty-five years, and I think this fits my experience. I do not believe I have ever met a socialist who approved of the results I think socialism would produce, have encountered few in recent decades who approved even of the results it did produce in the Soviet Union or Mao’s China. While the rulers of China since Mao have been reluctant to say negative things about him, what they replaced socialism with was “socialism with Chinese characteristics” — which is to say the same sort of mixed economy that elsewhere is called capitalism. Similarly, I do not know any libertarian who agrees with socialists about the material consequences of the alternative systems — if he did, he would find it hard to remain a libertarian.

The Economic Case for Liberty: A Short Sketch

All societies beyond the smallest face the coordination problem: To get almost anything done, to make an automobile or even a pencil, you must somehow coordinate the acts of a large number of individuals, arrange for someone to produce the wood for the pencil, someone to produce the chain saws to harvest the lumber for the wood, someone to produce the steel for the chainsaws, someone … . Add it all up down the branching chain and a pencil is produced by the coordinated acts of millions of people.

How do you do it?

The obvious answer, centralized control, someone at the top who figures out what everyone should do and makes him do it, works for small groups but scales badly. The more layers you have, in a firm or a socialist polity, the harder it is for the boss to know what is being done, figure out what should be done, make people do it. The farther the boss is from the people he controls, the less likely it is that his interest is the same as theirs, hence even if he succeeds in producing the outcome he wants it may not be the outcome they want.

The alternative is decentralized coordination. Divide the problem up so each individual solves a small part of it, makes decisions close to him, ones that depend on what he knows and what he can control. The trick to making this work is to somehow set things up so that the net benefit to each actor of a decision he makes is equal to the net benefit of that decision to everyone it affects, the sum of benefits net of costs.

That sounds impossible but an ordinary free market system comes surprisingly close to doing it. To produce something I hire workers, buy inputs. To hire a worker I must offer him at least as much as he could make doing something else or at least the value to him of his time spent in leisure, recreation or production; the cost to him of working for me is transmitted to me in what I must pay to get him. To buy an input I must either bid it away from someone else by offering him at least as much as it is worth to him or pay someone to produce it at least the cost to him of doing so; the cost for me to use an input is transmitted back to me through the price system in what I must pay to get it. I collect the value of what I produce when I sell it to a consumer. Any costs directly to me, my time and effort, are born by me. Hence all costs and benefits of my actions are transmitted back to me. If the sum is positive I will take the action and should, if negative should not and will not.

This is a very simplified sketch of the proof of the efficiency if the competitive equilibrium; the full version takes about two semesters of price theory.¹ Price theory is the core of economics, something any competent economist understands. So why don’t all of them agree with my politics, why are not all economists libertarian anarchists or at least libertarians?

There are several reasons, but the important one for economists is that what I have sketched is only a first approximation to a solution to the problem of decentralized coordination. Prices transmit costs and benefits back to the individual actor but they do not do it perfectly. Sometimes he takes an action that imposes costs on others without requiring their consent, so does not have to pay them its cost — emits air pollution from his factory, goes to a party with a cold or covid. Sometimes he produces a benefit but cannot control who gets it so cannot charge for it — a radio broadcast or repainting my house a lovely shade of blue. We call the former a negative externality, the latter a positive externality, call what keeps me from charging for a positive externality the public good problem, the problem of charging for a good whose producer cannot control who gets it.² In the former case his gain may be positive when ours is negative, so he does it and shouldn’t, in the latter case doesn’t do it and should. The government could correct those outcomes. That, what economist call the problem of market failure, is why most economists favor some level of taxation, subsidy, regulation, government interventions in the market to tweak it into doing the right thing when it would otherwise do the wrong, push it closer to coordinating to maximize the net value produced by our actions.

The argument is correct. Only the conclusion is wrong.

The mistake is thinking of the government as a wise and benevolent despot, a philosopher king, intervening to correct the mistakes of the imperfect market. The government is not a person, not something that can be modeled as a person. In a very real sense the government does not exist.

What we think of as the acts of a government are the result of the interacting actions of a large number of individuals, each acting in what he sees as his own interest, a political market like the private market but operating under different rules. Just as with the private market, to get it to produce the right acts, the acts that maximize total value to persons summed over everyone, it needs to be set up in such a way that each actor bears the net costs of his action, making his interest the same as our interest. The free market does that imperfectly, the political market much more imperfectly. Consider some examples:

The mechanism that is supposed to make a democratic government act in our interest is voting — if politicians do the wrong thing we vote them out. That requires the individual voter to know what politicians are doing and what they should do. That is not costless. No politician runs as the bad guy. No legislator titles his farm bill “A Bill to Make Farmers Richer and City Folk Poorer.” To vote rationally requires considerable time and effort, paid by the voter.

That is a cost. What is the benefit? The benefit to us if everyone does it is that we get better government. The benefit to the individual voter of doing it, of finding out who the better candidate is and voting for him, is an increase by perhaps one part in a million of the chance that the better candidate will win. Electing the better candidate benefits hundreds of millions of people but the individual voter is only one of them, pays all of the cost of his being an informed voter, collects a trivial fraction of the benefit.

Suppose an auto tariff benefits the winners, the auto companies and their workers, by ten billion dollars, costs the losers, purchasers of autos and producers of export goods — if we import less we will export less (details in previous posts) — twenty billion. The winners are a handful of companies and one union. Bribing congress to pass the tariff faces the public good problem since it benefits all of them, but producing a public good is not that hard if the public is very small. They agree that each will chip in, with dollars or votes, if the others do, can probably raise a billion or two to buy the tariff. The losers are several hundred of millions of individuals. There is no way they will raise even a fraction of that to pay the legislators not to pass the tariff.

There are numerous other examples. Actors on the private market usually pay most of the cost of their actions and receive most of the benefit. Actors on the political market rarely do either. On the private market, market failure is the exception. On the political market it is the rule.

That is a consequentialist argument sufficient, I believe, with more details filled out, to establish a strong presumption against shifting decisions from the private to the political market. It is not a proof that you should never do soMy web page, with the full text of multiple books and articles and much else

Past posts, sorted by topic

A search bar for past posts and much of my other writing. There might be situations where market failure on the private market would be so severe and its consequences so dire that it is arguably worth risking government action; the standard example, and the reason most libertarians are not anarchists, is national defense. But it is, I think, a sufficient argument to imply that all developed countries and all of their major political parties support substantially more government action than would be in the interest of their populations, judged by widely shared criteria.

---

My web page, with the full text of multiple books and articles and much else

Past posts, sorted by topic

A search bar for past posts and much of my other writing

1. Or my Hidden Order (free pdf), written as a substitute for a price theory course for people who prefer to teach themselves things.

2. The full definition is more than that but that is the part that matters here.