David R. Henderson  

Krugman's Insightful Analysis of Libertarians

Will the EU help Uber?... Krugman versus the Audit Heuri...

I was too busy on April 1 to post an April Fool's Day entry. This isn't one. But my use of the word "insightful" in the title is meant to be ironic.

A few days ago, Paul Krugman posted that "there basically aren't any libertarians." How did he reach that conclusion? The way he often does when he does a hatchet job: by coming up with a distorted criterion.

Actually Krugman quotes Nate Cohn saying that there basically aren't any libertarians. He then adds:

And that's true. I wish I could say that Rand Paul can't win because he believes in crank monetary economics, etc. But the truth is that these things matter much less than the fact that not many Americans consider themselves libertarian, and even those who do are deluding themselves.

So, in Krugman's view, we have only non-libertarians and putative libertarians who delude themselves.

A few lines later, though, Krugman concedes that there are "very few" libertarians.

At the end of his piece, Krugman comes very close to the more extreme version, writing:

There are almost no genuine libertarians in America -- and the people who like to use that name for themselves do not, in reality, love liberty.

Get it? There are "almost no" genuine libertarians in America. And "the people," not "some people," who "like to use that name for themselves" don't actually love liberty. And he knows this because he reads us so often and so carefully.

The best answer I've seen to Krugman is by David Boaz, who actually is a libertarian, thus instantly refuting the initial extreme version of Krugman's claim. Boaz's piece is titled "Paul Krugman Can't Find Any Libertarians."

Boaz puts his finger on the problem with Krugman's analysis:

Part of the trick here is that Krugman has used a vague term, "socially liberal," for one of the dimensions of the matrix, and a radical policy position, "no social insurance," for the other dimension. The logical way would be to use either common vague terms for both dimensions - say "socially liberal/conservative" and "fiscally liberal/conservative" - or specific and similarly radical terms for both dimensions, something like "no social insurance" and "repeal all drug laws." Wonder how many people would be in the boxes then?

"No social insurance" is a very radical position. Even many libertarians wouldn't support it. Like Hayek. So to find the divisions in our society, we might choose a specific issue of personal freedom - gay marriage, say - along with an equally controversial economic policy such as school choice or a constitutional amendment to balance the budget.

If you use either set of dimensions, it's pretty clear that you're going to get substantial numbers of libertarians (broadly speaking) and also of those incongruous creatures Krugman can only call "hardhats." Libertarians often call people who support substantial government intervention in both economic and personal issues "statists" or authoritarians. In our studies we call them populists, as does the Gallup Poll.

Boaz has a number of other insights also, and I recommend reading the whole thing, which isn't long.

I do want to add a criticism of Krugman, though, that I wish David Boaz had made. That is that the two-dimensional chart used to ferret out whether someone is a libertarian or something else should really be three-dimensional. Here's how I put it in a 2006 speech to a Libertarian Party audience:

I'm sure that many of you are familiar with the libertarian-designed "World's Smallest Political Quiz." Let me ask you a question: How many questions does that quiz have on foreign policy? [Someone in the audience answered, correctly, "Zero."] We libertarians have honed our principles and applied them to literally hundreds of domestic policy issues. We've done a great job. The depth of our understanding of how to apply our principles to these issues and of the importance of peace in the domestic realm is truly something for us to be proud of. But we haven't given nearly the same care to examining foreign policy.

I added [Brian Williams caveat--I think I added]:
You could advocate that the U.S. government drop a nuclear bomb on Canada and you could still be counted as a complete libertarian.

Comments and Sharing

COMMENTS (33 to date)
Jon Murphy writes:

With respect, it's not just Krugman who does this, but many commentators on both sides.

I have long thought about why this is and I think it's because nobody quite knows what to do with us. We don't fit in easily into some pre-checked box.

Besides, it's always easier to deal with the radicals than to deal with the mainstream. Folks would much rather attack Elizabeth Warren's ideas than a Blue Dog Democrat. Or attack Rick Santorum rather than Scott Brown.

David R. Henderson writes:

@Jon Murphy,
We don't fit in easily into some pre-checked box.
David Boaz, in the piece I cited, would certainly take issue with you.

Charlie writes:

There is a little bit of a contradiction in a wide definition of libertarian and arguing that libertarian policies are rarely tried. How many "libertarians" broadly defined are against open borders or even much more moderate immigration reform.

I meet many. Just look on Bryan's twitter feed. I tend to think the narrower description is more useful.

Jon Murphy writes:

@David R. Henderson

David Boaz, in the piece I cited, would certainly take issue with you.

I must not have expressed myself correctly. Let me try again:

Libertarians, in a very general sense, tend to favor issues that cross ideological boundaries. For example, we tend to support legalization (a typically liberal position) because of our belief in limited government. Under that same belief of small government, we tend to support reduced government spending (a typically conservative position). Because of this, commentators, such as Krugman or O'Riley, aren't sure how to handle us based on the typical Right/Left paradigm, so they tend to deal with the extreme views of some members of our group.

I mean, Democrats are typically left-wing and Republicans are typically right-wing. There may be some who are relatively more left than right, but generally speaking. Where on the political scale do Libertarians fall? We're not really left or right wing. We tend to be both. I consider myself a left-leaning libertarian. There are others who consider themselves right-leaning libertarians.

The tl;dr version: we don't fit nicely into the left/right spectrum.

Zeke5123 writes:

Krugman reads like a parody. He is pretty openly committing the "no true Scotsmen" fallacy.

I personally am opposed to our current social insurance scheme and am quite skeptical of most. However, those who want some kind of social insurance are not therefore non-libertarian. They just are a bit less libertarian. It is a sliding scale.

If Krugman ran his same poll but asked about minimum wage laws or rent control or free trade, I'd imagine the results would be different. Or put another way, people hold many views, some of which are contradictory. Expecting 100% fealty to a philosophic ideal seems foolhardy if you are being genuine. But if people hold generally to the principles of a certain philsophy, it is okay to call them a follower of that philsophy, even if they aren't platonic ideals.

Daniel Kuehn writes:

David - A major problem with Boaz's argument is that Krugman puts conservatives in the "no social insurance" box, so Boaz is clearly overstating what is meant by that when he says that it's so radical most libertarians wouldn't be in it. Of course if Krugman doesn't define it better we just have to take a stab at what he meant, but it is a blog post.

Personally I appreciated that Krugman framed it in terms of a policy position around social insurance (although he could have used others) rather than than in terms of "economic liberty" or even "fiscal conservative" or something like that. A lot of definitions of libertarianism beg the question by presupposing that libertarians are more supportive of liberty than non-libertarians. Krugman's formulation surely could be improved (and of course he could empirically be wrong) but at least it wasn't question begging in the way you often see the Nolan chart.

I like your third dimension.

Andrew_FL writes:

I don't like it when people refer to themselves as "Fiscally conservative/Socially liberal" and I like it even less when this annoying construction is said to be synonymous with libertarianism.

For one thing it makes very little sense to me to treat politics as divided into "social" issues and "fiscal" issues. A number of issues relevant to whether one is a "libertarian" in particular are completely absent from such an axis.

But "fiscal conservative" is probably the worse half of this construction. It's very nearly content free, seeing as it essentially means, being charitable here, being in favor of a balanced budget. Well you could be in favor of the government nationalizing the entire economy and be for a balanced budget.

By contrast, "Social liberal" is just lumping together a lot of issues that don't belong together. Which is still annoyingly bad, but less bad I guess. What's annoying is how it signals "I'm pro-abortion" but people who call themselves that always want to argue about pot and gay marriage.

To me, a test of where you stand politically would require 3 or 4 questions.

1. Should the government play a bigger or smaller role in the economy than it presently does?

2. Should the government intervene more or less in foreign affairs than it presently does?

3. Are you generally in favor of, or opposed to abortion?

4. Do you support or oppose some restrictions on firearm ownership/purchases?

(note that you may answer questions that seemingly offer binary options but exclude the middle ambivalently)

To be a "libertarian," in my mind, you must answer questions 1, 2, 3, and 4 as smaller, less, in favor, and oppose. If you answer the questions that way except for 3, and answer opposed there instead, you'd be what I'd consider a "right-libertarian" or a "conservative" (although you you could give an ambivalent answer to 2 or even answer "more" and if your answers were otherwise as I said, be a "conservative" as I would personally define the terms, you obviously would not be a "libertarian")

Of course, I recognize that not everyone will agree with this. In that respect I'd probably be more lenient on Krugman, actually. There's no reason he can't understand "libertarian" to mean whatever he thinks it should. Of course, it's not clear why he finds it useful to define the word to be a nearly empty category. Apart from finding that gratifying in and of itself, which he really shouldn't.

In case you're wondering my definition of "libertarian" is designed to exclude me, not empty the category.

Lupis42 writes:

With a little modification of the chart, I can use it to prove the Libertarianism of Paul Krugman. Behold:

****** Pro Trade w/ sweatshop nations | Anti
Social Lib Paul K, Libertarians | Dems
Social Con Republicans | Nobody

MikeP writes:

A major problem with Boaz's argument is that Krugman puts conservatives in the "no social insurance" box, so Boaz is clearly overstating what is meant by that when he says that it's so radical most libertarians wouldn't be in it.

I think you mean "a major problem with Krugman's argument."

Seriously, to a person, libertarians believe in less social insurance than conservatives. So if there are no libertarians because they must believe in no social insurance, then surely there are no conservatives either.

Daniel Kuehn writes:

No, I mean Boaz.

It's precisely because libertarians believe in less social insurance than conservatives and all of conservatives fall in Krugman's "no social insurance" box that Boaz's strong reading of Krugman has to be wrong.

"No social insurance" obviously has some threshold that not only includes the few libertarians Krugman thinks exist, but also conservatives. It clearly is not the sort of radical standard that Boaz suggests.

Krugman could define it better for his readers, obviously, but that doesn't mean Boaz's attempt at interpretation makes sense either.

Daniel Kuehn writes:

MikeP -
Let me put it this way. Krugman is, without a doubt, vague. Boaz is, without a doubt, wrong. Even with how vague Krugman is on what he means exactly, it's clear that Boaz cannot be providing a good guide to the post on this point.

MikeP writes:

Krugman is not vague. Krugman is delusional.

There ought in principle, you might think, be people who are pro-gay-marriage and civil rights in general, but opposed to government retirement and health care programs — that is, libertarians — but there are actually very few.

Change 'pro-' to 'anti-' and 'libertarians' to 'conservatives'. The statement has the same truth value.

This is also known as the "no true strawman" fallacy.

Daniel Kuehn writes:

MikeP -
I feel like you're making a different point here than what we started with. My point was that "no social insurance" can't be as strict as Boaz is reading it because Krugman's chart has conservatives not supporting social insurance.

But if you're just going to go around calling people delusional because you don't agree with them don't worry about continuing the conversation - I'm not interested.

Capt. J Parker writes:

@Daniel Kuehn,

"so Boaz is clearly overstating what is meant by that (no social insurance) when he says that it's so radical most libertarians wouldn't be in it"

Strongly disagree. Dr. K is not bashful about making the politics he disagrees with appear as radical as possible. If you had picked say, increase the federal budget deficit (Like DeLong is arguing and Krugman is cheering on) vs decrease the deficit as the policy position framing then his chart would indicate there were a lot of libertarians and not that many liberals. So, Boaz' argument is perfectly sound. Krugman can't find any libertarians because he wants everyone to believe libertarians all hold a radical straw-man policy position which they don't .

MikeP writes:

My apologies. You actually made a very good point that Boaz missed -- a point that results in a far stronger argument against Krugman's claim than Boaz made. Thanks for that.

And I don't use delusional because I disagree: I use delusional because, reading the article for third time, there is zero reason given for why the libertarian box is empty except for this faulty logic which must also empty the conservative box.

Daniel Kuehn writes:

MikeP -
He can be empirically wrong and not delusional. I'm not sure how I make a strong argument against Krugman's claim although maybe I'm not sure what you mean by Krugman's claim.

There's a part of the Krugman post I like and a part that I don't like. I like that the dimensions of his Nolan chart aren't question begging compared to a lot of definitions of libertarianism out there (not that there aren't remaining questions about it - I mean that it doesn't beg the question).

The part of the Krugman post I'm not so sure about is the empirical claim that there are few people in that square. Boaz doesn't help much because he's looking at something different. I find Krugman's own sense that a lot of people who talk libertarian are more conservative plausible but ultimately uncertain - I don't have a strong view on that.

Zeke writes:

[Comment rejected for name-calling. Email the webmaster@econlib.org to discuss editing your comment.--Econlib Ed.]

Hazel Meade writes:

It's obvious to me that the left is terrified of Rand Paul.

Their responses to his candidacy have ranged from critiquing him as not libertarian enough, to claiming he's not a threat because there aren't enough libertarians to vote for him. One seems to acknowledge the appeal of libertarianism as a policitcal philosophy, the other denies that anyone really believes. In other words, it's a comination of denial and disbelief.

Hazel Meade writes:


I think your four points might be a good way of categorizing the group of people who currently identify as libertarian, but I don't think it really works to define the modern libertarian movement.

To me a libertarian can be anyone who follows a philosophy that starts with individual liberty (self-ownership) private property rights, the rule of law, and equal justice, attempts to derive a comprehensive political philosophy from that. Depending on one's reasoning it is *possible* for people to come up with arguments against at least the latter 3 points that are grounded in libertarian philosophy. There are pro-life libertarians and hawkish libertarians and anti-gun liertarians. What unifies them is that they all attempt to justify their position based on some fundamental set of libertarian principles.

Miguel Madeira writes:

Krugman is clearly using "no social security" to cover also "reductions in social security" (much probably he believes that people who want to reduce social security secretly consider "no social security" as the optimum point).

Ann note that the central point of Krugman's post is almost "almost all enemies of liberalism are conservatives, even if they call themselves libertarians", then the reason could not be the "no social security" issue (that he atributes both to conservatives and libertarians)

David Henderson: "I do want to add a criticism of Krugman, though, that I wish David Boaz had made. That is that the two-dimensional chart used to ferret out whether someone is a libertarian or something else should really be three-dimensional. "

Why this is a criticism? He, en passant, also includes that dimension ("And positions on foreign affairs — bomb or talk? — are pretty much perfectly aligned too").

Andrew_FL writes:

@Hazel Meade-Judging from the way some libertarians have reacted to Rand Paul having positions they don't agree with, I'm not sure you're correct. In particular I think the modern libertarian movement is determined to consider "hawkish libertarian" an obvious oxymoron.

Which, actually, I'd be inclined to agree with.

I also find your definition, while appealing, to be operationally over-broad. It would cause many people not usually considered to be libertarians to arguably be included.

Pagna writes:

It is not surprising to hear this from Krugman. I really doubt if he even did read Free to Choose, not to mention the Constitution of Liberty.

Even he classifies himself as Keynesian, his Keynesian theories would collide with many other Keynesians as well.

R Richard Schweitzer writes:

This is not in support of any particular candidate's "viability."

If we are discussing actions in the context of voters "thinking" they are influencing the course of government, then:

"Normative Libertarianism is framed by the impacts of the functions of governments on Liberty and thus to limit those impacts by limiting those functions."

Extending, or attempting to extend, that normative consideration to all other human interactions that may impact individual liberty
does not change that normative consideration or the need for it. They stand separately.

We can have the normative, and it is widely shared, with or without all the other.

James writes:

Daniel Kuehn wrote, "A lot of definitions of libertarianism beg the question by presupposing that libertarians are more supportive of liberty than non-libertarians."

This is false. A definition does not presuppose any claim about the state of the world. It maps a label to a concept, in this case "libertarianism" to a political philosophy which favors the role of government in society. Does the definition of a knife presuppose a state of the world in which there is something to cut?

However, the position which Daniel claims is presupposed is worth considering. The reality is all decisions get made by someone. If a new Constitutional amendment precluded the government from forcing people to participate in social programs, that would increase my freedom (I could quit paying into Social Security) but it would reduce the freedom of politicans (they couldn't force me to participate). If the government got out of the marriage business, I would have the freedom to marry whomever I please, but politicians would lose their freedom to tell me who I can marry. Since many people expect politicians to do what they would have those politicians do, I can see why libertarianism might strike someone as unsupportive of the liberties they find important.

Daniel Kuehn writes:

James - you can replace "presupposes" with "defines" if that makes you more comfortable, but it still begs the question about the relationship between political philosophy and support for liberty.

It's certainly not a question of the liberty of politicians, it's a question of whether the activities of a government are a meaningful or exhaustive metric of liberty. This is what's so nice about Krugman's issue - it is much clearer on the ACTUAL difference between libertarians and non-libertarians: their view of the state.

Sean writes:

Daniel Kuehn: The issue is not "begging the question". Presupposing a definition of liberty or anything else is not begging the question. The fact is that libertarians may have a different definition of "liberty" than do liberals, but this does not involve begging the question. It is a matter of using a word to mean two different things. This is an old issue, and it was, I think, John Dewey who explicitly sought to redefine the word liberty to mean what libertarians would mean by "power". In doing this he simply used it in an equivocal sense, making it logically a different term. There is no question-begging going on by libertarians or liberals on this.

Tangential question: Does Krugman actually believe that there are no hardhats--no socially illiberal, anti-social insurance people? Has he never heard of National Socialism? Or does his argument depend solely on some alleged fact about the way things are in modern America? I thought he was making a claim that was quasi-psychological or typological and, therefore, universal.

Don Boudreaux writes:

Sean: You're correct about the confusion of the concept of liberty with that of power (or practical ability to achieve desired ends). Dewey (as I understand matters) was indeed influential in promoting this confusion. Alas, though, so too did some otherwise clear thinkers get caught in this trap - for example, Frank Knight, George Stigler, and Herbert Stein.

Freedom, as I understand and use the term - which is, I'm confident, how most other libertarians understand and use the term - is a condition of being free from the control of the state. Philosophers can, and should, nitpick this, and any other, such definition in order to inspire clarification. But as a practical matter, a person in society is more free the less is the ability of the state to override or to impose conditions on his or her choices (and vice-versa). A person who is free in this sense might be very poor and have his or her range of choices very tightly constrained by material circumstances. A person who is very rich, economically, might be less free than the very poor person if the rich person lives in a society whose government imposes more restrictions than does the poor-person's government on citizens' ranges of permissible actions.

My practical inability to buy a private jet in no way means that I am less free than is Warren Buffett, Bill Gates, or Tiger Woods. The confusion of "freedom" with "power" or "ability" is fundamental, and dangerous.

Greg G writes:


You say, "Freedom, as I understand and use the term - which is, I'm confident, how most other libertarians understand and use the term - is a condition of being free from the control of the state."

I think you are right that that is the main thing that separates the way that libertarians and non-libertarians understand freedom. And this is why I do not identify as a libertarian even though I believe I care as much about liberty as anyone else.

If I am victimized by a crime by a non-state actor that is just as much an offense against my freedom as if the state did it. It's not obvious to me why only state actions should count as offenses against freedom.

I think it is obvious that history shows freedom has flourished the most in the presence of a state that establishes the rule of law and the tradition of a judicial system that tries to protect everyone's rights regardless of their ability to pay. There are dangers to freedom on all sides, not just on the government side.

When central governments break down we do not tend to see a flourishing of human rights. We tend to see a return to tribal and kin based systems where the individual is seen as less important than the group.

MikeP writes:

Greg G,

Curiously, you have perfectly stated the libertarian case for the state: the state's sole reason for existence is protecting the rights of the individuals under its jurisdiction.

The difference between a crime by a non-state actor and an offense against freedom by a state actor is that the state should be preventing both, but the state is actually complicit in the latter. The latter is a matter of policy and therefore a political issue that demands a political position.

No one disagrees that the state -- if one exists -- should be preventing crimes by non-state actors. So there is little reason for libertarians to make a big deal about it: it's not an issue of contention.

The disagreement comes because some people believe that, say, taxing 18-year-olds to support millionaires, throwing individuals in prison for consuming substances, prohibiting residence and employment based on where someone was born, or drafting citizens to go kill enemies are crimes by state actors. Other people believe that these aren't crimes, but merely the things that we choose to do together.

Greg G writes:


You say, "No one disagrees that the state -- if one exists -- should be preventing crimes by non-state actors. So there is little reason for libertarians to make a big deal about it: it's not an issue of contention."

There is a lot you are sweeping under the rug there. Some (probably most) self described libertarians believe a state should exist to do only the things they want it to do. (By the way, the rest of us also want the state to do only the things we think it should do.)

Other self described libertarians think no state as we know it should exist. That is a huge difference. It (the issue of whether or not there should be a state) is a big deal and it is an issue of contention. Of course you are right that the idea that protecting rights is an important thing for the state to do is far less controversial.

It is not obvious to me that Don thinks a state should exist at all but he can clarify that if he chooses. Of course it's much easier just to dodge that issue and criticize the state.

James writes:

Daniel Kuehn:

I did not cite my discomfort as an objection to your earlier post. Please read more carefully.

"It's certainly not a question of the liberty of politicians..."

Literally everything about libertarianism deals with what decisions politicians should be at liberty to make. I'm sure you know this, so I'm at a loss as to how you could say otherwise.

MikeP writes:

It (the issue of whether or not there should be a state) is a big deal and it is an issue of contention.

Actually, it's not that contentious. Both minarchist and anarchist libertarians envision substantively the same society, and both have the same definition of liberty. The latter simply think that a state cannot reliably guarantee a libertarian society and that such a society does not require a state.

These are angels on the head of a pin debates compared to where we stand today. Both minarchists and anarchists want neither non-state actors nor state actors committing crimes, and both would find most of what government does today to be abrogating individual rights rather than securing them. And the question of whether Don is an anarchist or not is irrelevant in the question of whether the state's violating one person's individual rights in order to give another person more choices is liberty or not.

Greg G writes:


No matter how minarchist (spell check wants to correct that to monarchist which also reveals something about how popular these ideas are) you will be faced with a few anarchists who feel coerced by the activities of your minarchist government.

All possible arrangements leave someone feeling coerced. Constitutional democracy leaves the fewest people feeling coerced.

As for what is the correct definition of liberty, language is entirely conventional and those conventions are determined in the most anarchistic way possible. That is to say everyone gets to decide the meanings of the words they use and hear. The prevailing usage is the product of this anarchistic process. Be careful what you wish for.

Comments for this entry have been closed
Return to top