Bryan Caplan  

Krugman versus the Audit Heuristic

Krugman's Insightful Analysis ... The Public Choice Behind Carbo...
My two favorite liberals have repeatedly told me, "Paul Krugman is right about virtually everything."  When my favorite liberals praise Krugman's position about a topic I don't thoroughly understand, I hold my tongue.  Recently, though, Krugman staked out strong positions on a topic I know intimately: the ideological structure of public opinion.  And he's factually wrong.

Krugman notes that there are four logically possible combinations of social liberalism and economic liberalism.  He calls their adherents liberals (high on both), conservatives (low on both), libertarians (high on social liberalism, low on economic liberalism), and hardhats (low on social liberalism, high on economic liberalism).  Since social liberalism and economic liberalism are positively correlated, the latter two categories are relatively rare.  So far, so good.  But then he leaps to the claim that the latter two categories are absolutely rare - about as rare as they'd be if social and economic liberalism were perfectly correlated.  Krugman:

You might be tempted to say that this is a vast oversimplification, that there's much more to politics than just these two issues. But the reality is that even in this stripped-down representation, half the boxes are basically empty...

"Basically empty"?  Gallup, Pew, and the American National Election Studies all disagree.  Cautious definitions put the libertarian share of the U.S. population at 9-14%.  Broader definitions put the share much higher.  Libertarians themselves have done most of the data analysis, but you don't have to trust us.  Nate Silver concurs.  Or check out some social and economic questions in the General Social Survey for yourself, and see how they correlate.  For example:

Social Questions

1. "Do you think the use of marijuana should be made legal or not?" (GRASS, "Legal"=1, "Not legal"=2)

2. "Suppose this admitted homosexual wanted to make a speech in your community. Should he be allowed to speak, or not?" (SPKHOMO, "Allowed"=1, "Not Allowed"=2)

3. "If such a person wanted to make a speech in your community claiming that Blacks are inferior, should he be allowed to speak, or not?" (SPKRAC, "Allowed"=1, "Not Allowed"=2)

Economic Questions

1. "Please indicate whether you would like to see more or less government spending in each area. Remember that if you say 'much more,' it might require a tax increase to pay for it. b. Health." (SPHLTH, "Spend much more"=1, "Spend more"=2, "Spend same"=3, "Spend less"=4, "Spend much less"=5)

2. "Please indicate whether you would like to see more or less government spending in each area. Remember that if you say "much more," it might require a tax increase to pay for it. g. Unemployment benefits." (SPUNEMP, "Spend much more"=1, "Spend more"=2, "Spend game"=3, "Spend less"=4, "Spend much less"=5)

3. "Generally, how would you describe taxes in America today... We mean all taxes together, including social security, income tax, sales tax, and all the rest. a. For those with high incomes, are taxes..." (TAXRICH,"Much too high"=1, "Too high"=2, "About right"=3, "Too low"=4, "Much too low"=5)

Here's the correlation matrix.


Mind the coding.  On the social questions, lower answers are always more liberal.  On the economic questions, lower answers are more liberal for spending, but more conservative for taxes.  Social liberalism therefore statistically comes as a package, and so does economic liberalism.  The correlations between the social and economic questions, however, are much smaller, and some have the wrong sign. 

No doubt the low and irregular correlations partly reflect measurement error.  As I've emphasized myself, low correlations between individual issues don't disprove the one-dimensional model.  But even if you put responses on common scales and sum them to reduce measurement error, the correlation between social liberalism and economic liberalism is only .06.  If you put in a hundred questions, measurement error would largely wash out, so you might get the correlation up to .2 or .3.  But getting it up to .5 would be like pulling teeth.  Plenty of data miners have tried!  And with a .5 correlation, the libertarian and hardhat boxes remain very far from empty.

When I judge wide-ranging thinkers, I often use the following rule of thumb.  Call it the Audit Heuristic.

1. Read what they say about a topic I know very well.

2. See how reliable they are on that topic. 

3. Assume that they're about as reliable on topics I don't know well as they are on topics I do know well.

For the case at hand, Krugman fares poorly against the Audit Heuristic.  I don't expect my favorite liberals to change their mind about him because one of his posts is mistaken.  But my favorite liberals should at least admit that Krugman has given me reason to doubt his reliability.  And I'd trust my favorite liberals more if they were quicker to concede Krugman's specific flaws rather than sing his general praises.

COMMENTS (20 to date)
Shane L writes:

"Krugman notes that there are four logically possible combinations of social liberalism and economic liberalism. He calls their adherents liberals (high on both)..."

A small side point, and I hope it won't be distracting: I know there have been discussions before about the use of the word "liberal" to describe left-wing perspectives but I am totally surprised to see "economic liberalism" here to mean "economic illiberalism"!

Greg G writes:

>----" Cautious definitions put the libertarian share of the U.S. population at 9-14%."

I'm skeptical about how "cautious" that is especially for a term that is is defined in such a spectacularly wide variety of ways. I know a number of people who call themselves libertarians and proudly voted for George W. Bush twice. No one thinks of themselves as opposed to liberty. Some others consider themselves libertarian even while advocating that the government should force a rape victim to carry her assailant's baby to term.

The Libertarian party has run a Presidential candidate every year since 1980. Their very best showing has been 1% of the vote. This is an extremely poor showing even by the standard of other third party candidates.

ted writes:

[Comment removed at commenter's request--Econlib Ed.]

Brad writes:

I love the Audit Heuristic and use it myself. When someone is wrong about something I know a lot about I discount many of their other views that I don't know much about.

I cut folks some slack if thier views are explained without a high confidence level. But Krugman doesn't say ANYTHING speculatively.

AS writes:
The Libertarian party has run a Presidential candidate every year since 1980. Their very best showing has been 1% of the vote. This is an extremely poor showing even by the standard of other third party candidates.

You can't equate general election results to true preferences. In winner-take-all two-party elections, people tend to vote strategically rather than truthfully. Under proportional representation, you tend to get more truthful voting.

Since a very basic rational voter model predicts 0% voting shares for third-party candidates, I am actually surprised the Libertarian party can garner 1%. The true fraction of Libertarians must be much more than 1% to induce that many to "throw away" their vote.

CMOT writes:

Krugman's use of the term "hardhats" is a tell - it's a Vietnam War era term for reactionary construction union members who sometimes beat up hippies and war protestors. See here They were old school New Deal democrats who were in the process of becoming Reagan republicans.

He's locked into 1970 thinking, and his analysis might make sense in a 1970 framework, if all you had from them were headlines and not opinion research, but not now.

Maniel writes:

Well said; Ron and Rand Paul would agree.
A better measure might be the percentage of votes cast for Ron or Rand - who, in my judgment best represent the Libertarian perspective among high-profile candidates - in their respective quests for the presidential nomination by the Republican Party. Note also that each obtained enough votes, in Texas and Kentucky, to land them in the house and senate respectively.

Greg G writes:


I agree that third party candidates are at a big disadvantage compared to major party candidates and you have done a good job of explaining why that is so.

But why then do Libertarian Party Presidential candidates do so poorly compared to OTHER third party candidates? By the standards established by John Anderson, Ross Perot and even Ralph Nader, Libertarians are doing abysmally at the polls.

Andrew_FL writes:

That the #1 "Social Question" was about Pot and not Abortion goes exactly to the complaint I raised about the term "Social Liberalism" in a comment of David's post.

As can be seen from comments like Greg G's this is the real social issue that defines Social Liberalism. We're just supposed to pretend Pot is more important.

Hazel Meade writes:

Greg G,
I don't think it's fair to compare cult-of-personality third party candidates to libertarian party candidates. How well has the Reform party polled since Ross Perot stopped running for President? How well has the Green Party polled without Nader?

In the absence of a charismatic personality at the helm, the LP always does consistently better than any other third party, with the exception of maybe the Green Party in certain localities.

Greg G writes:


John Anderson and Ross Perot were virtually unknown to the average voter before they ran. Nader was better known.

If you think any of them was a "charismatic personality" then you have a very different concept of charisma than I do.

David R. Henderson writes:

@Greg G,
I find the different views of charisma interesting. I think Ross Perot was incredibly charismatic: “Larry, Larry.” And Nader pretty charismatic. I’ll give you John Anderson though.

Neil S writes:

Andrew_FL and Greg G,

I would argue that Pot is a much more appropriate choice than abortion, because the question of use of force against others is not arguable.

While the two of you obviously disagree, many others do believe that a fetus has value as an individual and that abortion is an act of violence against that individual. Opposition to abortions on those grounds is not inconsistent with libertarianism.

Thus questions regarding abortion are not particularly appropriate (in isolation) for defining a degree of social liberalism in this context. The legalization of drugs on the other hand does not raise questions regarding balancing rights of individuals.

Neil S

Floccina writes:

I do not find Democrats much more socially liberal than Republicans. I do not think for example that Democrats want drugs to be completely legal, rather they seem to want kinder less harsh punishment for users and sellers.
They keep pushing cigarette taxes higher. they just want the cigarette taxes less harshly punished. They also seem to want to limit certain foods but I am sure that they think that they can do this stuff without cracking heads but the Republicans will get power and say this is how you treat these criminals (black marketers of cigarettes for example.)

Greg G writes:


I take your point that a person can be a libertarian and still have moral objections to abortion.

I don't identify as a libertarian even though I believe I care as much about liberty as anyone else. The reason I don't is that the term has a fairly specific meaning which doesn't apply in my case. It seems to me that the one thing that most unites libertarians is a much greater fear of government than the average person has. I am afraid I can muster only an average fear of government.

Now it seems to me that the decision of a rape victim about whether or not to abort is very different from the decision whether or not to abort an ordinary pregnancy. It is the most personal of all decisions and the last thing I would expect a libertarian to want government to be mixed up in for that reason.

Mark V Anderson writes:

I am only kind of a libertarian, but I have never understood why some think abortion is ok for a rape victim even if not for others. The fetus no more deserves to die in the case of rape than with consensual conceptions.

Using abortion as one facet libertarianism sounds nuts to me. The question is between those who believe that fetuses have zero rights and only develop them when born, and those of us who disagree. What has this to do with libertarianism? I don't even see why this is a question of "social liberalism," other than it has been adopted by left wing feminists as a signature issue.

Greg G writes:


I would say that both the fetus and the mother have some rights and those rights can come into conflict with each other. This is most obvious in cases where the pregnancy cannot be safely carried to term for medical reasons.

Rape is not different because the fetus is more deserving of death. It is different because the position of the mother is so different in ways that should be obvious. She did not voluntarily assume the risk of pregnancy in the way that a woman does who has consensual sex. She has concerns about her baby inheriting criminal tendencies that are different than would be the case if the father was not a rapist. She has been traumatized once and may be traumatized again by the knowledge she is carrying her rapist's baby especially if she is forced to do so.

Despite the fact that many disagree, it seems obvious to me that the relationship between the rights of the fetus and mother are constantly changing as the pregnancy progresses. If full human rights are present at the "moment of conception" then there is such a thing as a one celled human being with the same rights as a fully formed human. In any event conception is a process with many opportunities for things to go wrong, not something that happens in a "moment."

Using the abortion issue as the only thing defining libertarianism is indeed nuts as you say. Using it as one thing defining libertarianism is unavoidable. In my experience, libertarians are uncomfortable with the idea that some rights can conflict with other rights creating complex moral dilemmas. This is most obvious when we talk about a fetus or a child. So libertarians usually prefer to focus on other issues.


This seems flatly inconsistent with both what you yourself have written about this issue in the past and what I've seen from other sources.

You've previously said people's political opinions basically "boil down to roughly one big opinion, plus noise." If that's true, it's highly misleading to go about trying to identify libertarians based on their opinions on a small handful of questions (or, worse, a single question for each dimension, which is what some of the sources you link to do).

Furthermore, from what I understand attempts to identify a distinct libertarian cluster among the voting public have found this difficult.

I can understand you disagreeing with Krugman about what this means. But the phenomenon seems hard to deny.

LD Bottorff writes:

Imagine that the libertarians became focused enough to put their energy into a lobbying organization instead of a political party. What issues would they agree to lobby for? I can see that nearly 100% of those who consider themselves libertarian would agree that we should lobby for legalized pot. Are there any other issues on which a Liberty Lobby would agree? We have seen here that people who consider themselves libertarian don't agree on abortion. I suspect there are plenty of libertarian leaning folks who are concerned with unfettered immigration. Others see that anti-discrimination laws are anti-liberty. There is plenty of diversity of opinion in the crowd that considers itself libertarian.

Andrew_FL writes:

@Neil S-You misunderstood me, I'm pro-life. I was saying this is the main reason I don't, without qualification, call myself a "libertarian" and would never call myself a "Social Liberal."

Comments for this entry have been closed
Return to top