Rand Paul’s entry into the presidential race raises an interesting question: Just how many people consider themselves “libertarian” and what is the potential voting strength of this segment of the population?

Paul Krugman weighed in early with the brash claim that “there basically aren’t any libertarians.” But as David Henderson reminds us, once again Krugman’s opinions are nowhere near factual reality.

Gallup, which has been polling on the issue for quite some time, finds that 38 percent of the public identifies as “conservative” and 24 percent as “liberal” in the latest poll.

Gallup doesn’t ask people if they are libertarian, but fortunately other polls have asked that question in various ways. In 2006 the Cato Institute commissioned Zogby International to ask 596 voters this question: “Would you describe yourself as fiscally conservative and socially liberal?” Fully 59 percent of the respondents said “yes.”

Then Zogby asked the same question of the same number of voters, but this time they added the “L” label: “Would you describe yourself as fiscally conservative and socially liberal, also known as libertarian?” David Boaz recounts the results:

The addition of the word “libertarian” clearly made the question more challenging. What surprised us was how small the drop-off was. A healthy 44 percent of respondents answered “yes” to that question, accepting a self-description as “libertarian.”

If that result is anywhere close to accurate it means that libertarians out number both conservative and liberals among the electorate. Or perhaps more precisely, more people seem to lean libertarian than lean conservative or liberal.

So why don’t we see more elections in which a libertarian candidate is opposed by an anti-libertarian (someone who advocates more government both in the boardroom and the bedroom)? In other words, why aren’t more candidates consistently in favor of more government across the board, or less government across the board?

Think of the process of producing votes as analogous to a business producing a good or service. Just as a firm needs both capital and labor, so do political candidates. Then we can apply the:

Goodman/Porter Theorem on Political Coalitions: Candidates who are capital-intensive will seek labor-intensive supporters to add to their coalition; candidates that are labor-intensive will do the opposite.

This follows from the law of diminishing returns. The more capital you have, the less valuable one more unit of it is. Similarly, for labor.

Suppose that we start out with a Republican candidate who is supported with heavy capital contributions from the business community (because of her generally free enterprise positions) and a Democratic candidate who has a lot of support from labor (because of his support for various kinds of government intervention). Although, unions do give a lot of money to Democrats, they also provide a lot of boots on the ground, which I will call “labor.” So initially, the Republican is relatively capital intensive and the Democrat is relatively labor intensive. At this point a unit of labor is far more valuable to the Republican and a unit of capital is far more valuable to the Democrat.

Where can the Republican get labor? From social conservatives. Where can the Democrat get capital? From wealthy liberals who care about social issues (and the environment) and from The New York Times and other media outlets who believe in laissez faire in the market for ideas — no matter how regulated the market for goods and services is. I am counting newspapers editorials and biased reporting as a capital contribution — the alternative to paid advertising.

This theorem seem to fit the facts reasonably well. At least better than alternative explanations which assume that the process is basically irrational.

Political scientist Hans Noel argues that parties have, in a sense, been captured by ideologues who have persuaded those who pay close attention to politics to adopt “conservative” or “liberal” ideologies. Jonathan Bernstein at Bloomberg argues that:

[L]ooking for ideologies among voters is sort of pointless. People have impulses on various issues, but it is the parties that organize those impulses into something resembling overall world views. The parties do this by teaching their adherents which positions they should take on all those subjects the rank-and-file voter doesn’t care about much.

In other words, parties have ideologies, but not people? I don’t think so.

Back to our original question. Just how many libertarians are there? Krugman is joined by polling expert Nathan Cohn and by columnist Nate Silver in the belief that they are few and far between. But that depends crucially on how you define libertarian and how you word the questions put to votes. Political scientist John Sides, for example, thinks that a libertarian is someone who wants government to provide “fewer services.” But who wants fewer services? That’s not the public policy option. The question is: do you want more services plus higher taxes to pay for them or do you want fewer services and a tax cut?

In Krugman’s view of the world, the choice being a libertarian and not being one is a choice between “social insurance” and “no social insurance”? But what does “no social insurance” mean? Abolish Social Security? Abolish Medicare?

The question should be: Do you want to participate in a Ponzi scheme in which you have no court-enforced right to any future benefits or would you like private ownership of your retirement benefits, fully protected by the laws of contract?