Bryan Caplan is a professor of economics at George Mason University, senior research fellow at the Mercatus Center, adjunct scholar at the Cato Institute, and frequent contributor to Freakonomics. He also publishes his own blog, EconLog, which was ranked the sixth best “general economics blog” by Intelligent Economist (2018) and tenth most popular economics blog by Feedspot (2018). Mr. Caplan’s book, “The Myth of the Rational Voter,” was deemed “best political book of the year” by the New York Times in 2007. He recently published “The Case Against Education,” a deeply provocative and highly regarded treatise on how to fix the education system.
The Politic: I asked the same question to your colleague Tyler Cowen, but if an asteroid were about to hit New York City, how would you convince me to die a libertarian?
Bryan Caplan: How would I try to convince you to have no government response to the asteroid headed for New York City? My view is not absolutist, so the question in my mind would just be: is it feasible for private charity to take care of this? If not, it’s a small amount of money to save a lot of lives. That would be a very convincing exception, so I probably wouldn’t try to convince you in this scenario. I guess the other possibility is that nothing we can do is going to stop the asteroid. That would be another case where I’d say there’s nothing we can do, so no point going and making people contribute to something that isn’t going to work.
So, your belief in libertarianism basically comes down to whether it’s more practical for either private individuals or the government to act?
Well, there’s more to it than that. Michael Huemer’s “The Problem of Political Authority” is really the book that best explains what my views are. What it says is that libertarianism is a presumption: there’s got to be a high probability of accomplishing great good if you go and bend the usual rules about leaving people alone. The best explanation is probably this famous thought experiment where you’re a doctor with five different patients, each of whom needs a different organ, and then a guy wanders by with no family. Should you murder this guy to save five people?
This is one where I would say, as most people would, that you shouldn’t [murder him]– not to save five people. But to save a million people? That case at least seems a lot more likely to be right. But again, if it’s just a matter of killing one person to save 1.01 people? That would seem to be unjustified. I think this is a good description of actual moral reasoning, where people usually start with presumptions, and then the presumption is something that is hard, though not impossible, to overcome.
Would you then say that your belief in libertarianism isn’t deontological?
Here’s the thing. In philosophy, usually the deontological position is anything other than pure consequentialism. In other words, the consequentialist view is that you should go and kill 1 person to save 1.01 people, and anything that says not to do that would count as deontological. So, normally, deontological just means anything other than pure consequentialism. Of course, you could also view it on a continuum, and there I would say that I’m somewhere in the middle between the pure consequentialist view and the pure deontological view.
Under your view, how do you establish the point at which the government ought to intervene, or maybe in your words, overcome the presumption to leave people alone?
Right. Well, I would look at whether there’s strong evidence that we’re going to get at least a 5-to-1 benefit-to-cost ratio. There are two things. First of all, is the evidence strong or is it just speculation? Second of all, is it clear that you’re getting a large gain relative to the cost? That’s one where, given the uncertainty of the world and just how bad people’s predictions about political effects are, it’s a very hard hurdle to overcome.
I have this blog post on pacifism where I just say look– since action will always involve at least negligently killing large numbers of people, at minimum to be justified you would have to have a high degree of confidence that a war will actually do a lot more good than harm. That’s where I say that if you just take a look at how bad people’s forecasts are of the actual likely effects of war, it’s strong enough to really say that you shouldn’t participate in almost every actual war. Again, this isn’t a purely consequentialist view, but it’s also one not saying it’s impossible [for war to be justified]. It’s just saying that, once you put the burden on the person that wants to go and negligently kill a bunch of innocent people, then it’s very hard in the real world, given the difficulties of forecasting, to ever actually justify that. Meanwhile, the pure consequentialist might say, “we don’t know how it’s going to turn out, but it might be really good… or it might be really bad, so let’s throw the dice, kill a bunch of people, and see what happens.” You might even think of this as the honest argument for the last Iraq war, which, when you put it that way, doesn’t sound very good.
So, essentially what you’re saying is that when we take an action, there are often really large consequences to taking that action. We’re often very bad at forecasting those consequences, so in cases where we’re uncertain or there’s not enough evidence, we should always default to not taking that action?
Right. There’s that, and then there’s also the 5-to-1 standard as in that thought experiment involving the murdering of one person to get his organs to save five people. Almost everyone says you shouldn’t do it. So, you should have a similar standard for any government action of making sure you actually have strong evidence that this is going to do so much good that it will justify violation of the presumption of liberty– of leaving people alone.
And if you just think about daily life, when is it okay to steal? Is it okay to steal an item from someone because you’ll enjoy it more than they would? The normal view is no. What about to save somebody’s life? Yeah, in that case it probably is okay. The rule against theft is a lot stronger in consequentialism but still not absolute, which seems to be the more senseful view.
Again, that’s one of the main themes of the Michael Huemer book, “The Problem of Political Authority.” So, when libertarians talk about the non-aggression axiom, I say, “no– the non-aggression presumption.” It’s not something where you should do it no matter what, but it’s something where, unless you’ve got strong evidence indicating that you shouldn’t do it, then you should do it.
How does the presumption of liberty apply to situations in the status quo where we’re already taking an action – for instance, with Social Security benefits – where you might instead be repealing something?
Right, so the presumption of liberty is not the presumption to favor the status-quo. That’s a different presumption, which you could also have, and probably a lot of people also do have. I say the presumption is not in favor of continuing to do what we’re doing, but rather in favor of leaving human beings alone. Whatever presumption you choose is going to heavily tip the scales in favor of where your presumption is, and I say the presumption should be in favor of leaving people alone– not keeping the status-quo constant.
Again, this fits the intuition of most non-libertarians to a higher degree. For example, a law might say it’s illegal to be gay. On the one hand, the law hurts some people. But, it also probably solves some other problems, and a lot of people seem to like it. So, you could have the presumption to just stick with the status-quo. Or, you could have the presumption that people should be left alone, and that gay people should be able to practice their desired lifestyle. So, it’s the law that has to be justified, not the repeal of the law. That’s where I always start.
I’m often thinking about libertarianism’s ability to confront issues of racism, prejudice, and bigotry– things that people often don’t think can be sufficiently addressed through a libertarian philosophy. How do you, as both a libertarian and an economist, deal with these issues that might be less quantifiable in nature?
I’d say that a lot of these things are highly quantifiable. For example, what is the effect of racial discrimination on the wages of blacks? This is a question that economists have been measuring for forty to fifty years. And the strong consensus view is that, at most, a small fraction of the black-white earnings gap can be explained by racial discrimination. The way this research works is that you go and try to find white and black workers that are similar in every other way except their race, and then you statistically try to figure out how much of the gap can be explained by just their race. Of course, people do the same thing for gender and a bunch of other things.
What the actual research says is that only a small fraction of the gap can be explained by racial discrimination. And there’s actually a background theory about why that makes sense, which is that if it were really true that black workers were heavily underpaid compared to white workers, then why not just fire all of your white workers and replace them with black workers? So, there’s both theory and actual data in favor of this view that, in labor markets, the problem of racial discrimination is greatly overblown and most of the gap is not caused just by animus. Rather, there are actually differences in the job performances of different groups for many reasons. This is one case where I’d say it’s not a hard call because when you actually understand the issue, you realize that the markets have actually already taken care of almost all of the problems that people are complaining about.
Of course, if the problem is just inequality, then that’s one case where I would say there’s all sorts of inequality, so why is that actually something that is so bad the government needs to do something about it? Again, it’s different if you have people starving to death versus not being happy that they’re getting paid 5 percent less than what they ought to be getting paid.
So, I take it you don’t think inequality is intrinsically bad. But if it’s the case, for instance, that there are disparate mental health outcomes for people due to inequality, would that then compel government action?
It’s at least worth thinking about. I would say that the category of ‘mental health’ is so broad – and a lot of it is, in my view, a power grab to medicalize everything – that it’s just a really hard question to ask. For example, I have a very old-fashioned view on alcoholism as being a moral vice, and I think that people who are medically labeled as ‘alcoholics’ are perfectly capable of stopping drinking. If you want to characterize that as a mental health problem, then I would say the category is so broad that it includes almost everything. If you were going to put that in, then why not just put racism as a mental health problem? Or almost anything really?
If it’s something more along the lines of mental retardation or dementia, something that I don’t think is people just medicalizing behavior that’s a choice – and just to be clear, I don’t think those are like alcoholism – then as to whether the inequality is actually contributing to those is, again, do we have very strong evidence of a very large effect? And is there a realistic and highly effective remedy involving government?
If we don’t have all those things, then I’d say government shouldn’t do anything. It should leave people alone, which is, again, a presumption that people have in so many ways. If you just think about anyone you know and all the ways you could criticize their life choices, there’s barely a person I know where I don’t think that I know better than they do and what they should be doing with their life. And yet, like most people, I also say, “well, you may be messing your life up, but it is your life.” It’s not my place to do anything about it, except perhaps in very extreme cases.
That raises another difficulty people have with libertarianism, or at least their perception of libertarianism: that, for instance, you must respect an individual’s personal choice to smoke cigarettes. Yet, on the other side, some people might say that a rational actor would never choose to smoke cigarettes, which is perhaps evidenced by the fact that most people who do so regret their behavior afterwards. What do you make of that?
What I think about that is, first of all, that it’s totally consistent with the logic of libertarianism to say that cigarettes should be legal, but that I still oppose them and I’m going to try to persuade people to stop using them. So, those two things are perfectly compatible. I have the additional view that, besides not having government do things, I generally think that it’s not my business to go and tell people what to do with their own life, unless, again, you’ve got very clear cases. The easiest one, of course, is where there’s just some information they don’t have. You’d say, “well, you may not be aware that smoking cigarettes takes seven years off your life.” If it’s just a matter of sharing information, then I don’t think there’d be any issue. If it’s a matter of going and haranguing people – part of libertarianism is the free speech to harangue others – I would also just say, as my own additional ethical view, that I don’t harangue others except under very strong circumstances, or if they’re someone I know extremely well: if my kid were smoking, then I’d likely be a lot more naggy. Although, even there… if they’re an adult and still smoking, there would be a point where I’d be like “this is terrible, but if you ever want to talk about this or try and figure out how to stop smoking, I’m happy to come and help you.” Still, I would probably just eventually say, “you’re an adult– it’s not my place to go and tell you what to do.”
In terms of people regretting smoking, of course, people also say they regret heavy drinking. This is one where my view – again, I’m telling you my view, not the canonical libertarian view if there is such a thing – is that when people talk about regretting their habits, a lot of this reflects what psychologists call ‘social desirability bias,’ whereby people say things that sound good rather than the ugly truth. So again, what sounds good? I really want to stop drinking because I care about my family, but I just can’t help myself. What’s true? I prefer a beverage to my family. I think actions speak louder than words. And furthermore, society – while there are social punishments for heavy drinking, whether or not you say you’re trying to change – treats the person who very blatantly says, “I am drinking this beverage because the beverage means more to me than my family” far more negatively than someone who says, “poor me, I really want to stop.”
My view is that there are probably people who would stop if they’d have no cravings, but that’s very different from wanting to stop. There’s wanting to stop if I could just feel totally differently: “I would love to exercise every day if only it didn’t hurt.” But it does hurt, so in the real world, I don’t want to exercise every day because it’s unpleasant. Imagine if someone were getting on my case and saying, “why don’t you exercise more?” and I said, “I really want to but I just can’t do it.” By analogy, you invite your friend to a party and they say, “I can’t come.” Now what does that actually mean? Well, it’s possible they’re in Mongolia, and there’s no available form of transportation that would get them to your party at the time of the party– in which case it’s true that they can’t come. Or, it might be that they don’t want to come because they have something else that they’re doing, and the polite thing to say is “I can’t come” rather than “I choose not to come” or “I’ve got something better going on.” And I think that when people say, “I can’t stop drinking or smoking,” it’s very much like “I can’t come to your party.”
Wrapping up, what do you think is the biggest challenge to libertarianism?
I’m torn. I’ll give you two answers. One answer is just that people are so tribal. Almost everyone – about 90 percent of people in the U.S. – feels either committed to the red tribe or the blue tribe. I know that there are a lot of Independents, but when political scientists look at these Independents, most of them are still actually very Democratic or Republican. Only about 10 percent of the U.S. population is genuinely Independent. So, anyways, people have these very strong political identities, their whole social circle is tied with them, and this is the way they think of themselves. To become a libertarian would basically be like leaving your religion for a Mormon or something like that: it’s very personally painful and very socially painful.
The deeper challenge is just the highly emotional, cognitive style that most people bring to politics. For almost any political issue, when you really pay attention to what people do, it seems to me like the main thing is they have an emotional reaction– rather than doing the responsible thing of saying, “I have this emotional reaction, so I have to exert strong control, calm down, and consider arguments and evidence without being biased by my emotions.” That is not the way that almost anyone approaches political issues. Instead it’s like, “I have an emotion, and now I’m going to immediately act as if the emotion reveals the truth of the world.” So, I think this highly emotional style that people bring to politics… it’s one, of course, you can see people using almost everywhere. But, in daily life, on the one hand, when there’s the emotional reaction of “oh, what a cute little unicorn– I really want to buy it,” there’s also the “oh, but it costs $100.” So, for day-to-day decisions, people are pretty emotional all the time, yet there’s still a counterweight of “well, calm down and think– is it really going to be worth the costs?”
In politics, I see almost none of that counterweight; people are just looking for answers that are emotionally appealing. I think about my dad who’s a very committed, conservative Republican, and I’ve actually shown him an article and heard him say, “doesn’t seem like anything I’d agree with!” And it’s like, so you only read things you expect to agree with? How can you ever learn anything if you only read things you expect to agree with? But again, that’s sort of the way he approaches the issue, which is step one, does it say what I want to hear? Maybe I’ll read it if it does. Otherwise, it’s just trash. I don’t read that stuff. Whereas, again, the actual way you learn is by deliberately going and exposing yourself to things that you’re going to disagree with. Otherwise, how can you discover anything new?
This is all very closely tied to that social desirability bias that I was telling you about. Someone says, “we’re going to pass a minimum wage to help the poor,” and the socially desirable answer is to say, “oh, wonderful! Oh, that’s great!” You’re not supposed to ask, “will it really work?” or “what about the disemployment effects?” Those are not comments that people, in almost any society, would look upon with pleasure. So, it takes effort to actually think like a libertarian, it’s painful for you as an individual because you don’t get to just go with the flow and believe that the world works the way you want it to work, and it causes friction with other people who you’re upsetting by challenging the effectiveness of what they’re doing.
Again, if you see parents with a very sick child going and giving the kid some totally ineffective treatment, it’s hard to go and say, “what you’re doing won’t work” when they say, “look– we’re trying to save our child’s life.” If you know something better, at least you can say there’s a better way, but what if there is no treatment for the child? What if all you can say is “look– you can’t save your child… but at least you can avoid encouraging these charlatans from ripping people off and giving families false hope?” Just think about the way people look at you when you say that kind of thing. What you say could be 100 percent true, and yet people would respond very negatively. I think this is a lot of why libertarianism does so poorly in terms of persuasiveness: it’s not something that encourages people to go with their emotional reaction. It encourages people to delay emotional reaction until after analysis, and often that means that you just have to suppress the emotional reaction, which is not fun for most people. You know, I’d say you can get in the habit of it where eventually your emotions may realign, but that takes a long time, and some may never realign.