Does Bryan Caplan believe in free trade?
I may do a longer post on utilitarianism on Sunday, but since a recent post by Bryan Caplan perked my interest, I thought I would get a brief head start today. The context was a debate between Bryan Caplan and Robin Hanson on liberty vs. efficiency. Although Robin was defending efficiency and not utilitarianism, Bryan’s argument in his blog post is exactly the sort of argument that many philosophers make against utilitarianism, so I will respond on that basis. I should say that just as with my defense of the efficient markets hypothesis, I am not so much pro-utilitarian, as I am unimpressed with arguments against utilitarianism. Indeed, on Sunday I will express some of my own reservations with utilitarianism.
As with the EMH, I will rely somewhat on the pragmatist maxim “that which has no practical implications, has no philosophical implications,” but the specifics of my argument will be different. One of the most common strategies of the anti-utilitarian position is to assume some societal set-up which shocks our sensibilities, and then assume that it would satisfy the utilitarian criterion of maximizing aggregate happiness. Thus we might be asked to imagine a scenario where the total pleasures of the slave-owner exceed the suffering of the slaves, or where a very rich person derives much more utility from an extra dollar than does a very poor person. Would we still be willing to apply the utilitarian policy criterion? Bryan has an even more shocking example where the benefits to Nazi’s from the Holocaust exceeded the suffering to the Jews. (BTW, he used monetary criteria, so it is possible that he wasn’t thinking of an anti-utilitarian argument. But that sort of argument is often used against utilitarianism.) At the end of these thought experiments we are told that unless we are willing to embrace the society envisioned in the thought experiment, we must, on logical grounds, give up on utilitarianism.
I have several interrelated objections to this style of philosophical inquiry. I’d like to start with Richard Rorty’s assertion that the narrative arts (novels and film) produce liberal values. BTW, Rorty defines liberals as people who believe cruelty is the worst thing we do. Sunday I will try to equate utilitarianism with liberalism, broadly defined. So if Rorty is correct, how do we know that slavery was so awful? Because we have been exposed to accounts of slavery in the arts which vividly showed how the suffering of slaves was immeasurably greater that the frivolous pleasures of the slave-owner. Can we then turn around and use an imaginary slave-owning society that passes the utilitarian test as an argument against utilitarianism? I’m not sure that we can, unless one can show that our initial visceral reaction against slavery is based on non-utilitarian grounds, i.e. based on some abstract philosophical principle. And that’s much harder than many people might imagine.
My second objection is that if these thought experiments are useful, it ought to be possible to find some real world example, in some country, at some point in history, of an actual public policy that clearly would pass the utilitarian principle, and yet which shocks our moral sensibilities. And I’m not sure that it is possible. (I’ll consider a few likely suspects on Sunday.) If not, what’s wrong with using utilitarianism as a policy compass, on purely pragmatic grounds?
My third objection is that I don’t think Bryan has a plausible alternative procedure. Consider the question in this post’s title. I presume the answer is “yes,” but I didn’t want to be presumptuous. What really interests me, however, is the way Bryan would reach this answer. In the unlikely event he reads this post I’ll throw out two alternative procedures. To make it interesting, let’s start with a fairly routine case.
Assume that imports of Barbie dolls will eliminate 1000 jobs in South Carolina. Each job loss will have a devastating effect on the lives of the former workers. We’ll put a monetary value on the loss at $100,000, but keep in mind that these costs also involve things like divorce, alcoholism, etc., and not just foregone wages. Set against these losses, 60 million Americans will be able to go out and buy dolls at a “China price” that is $2 lower than the U.S. price. How do we go about determining whether free trade is the way to go?
1. Do a cold-blooded cost/benefit analysis. (Note, I understand that in the real world these would be done on broad principles, not on a case by case basis, but it wouldn’t change the logic of the argument.)
2. Go with our moral intuition. Which outcome seems worse? Which outcome do we find more intolerable to think about?
I don’t know about Bryan, but if I used the second procedure, I might end up on the anti-free trade side. Now even I wouldn’t claim that my little thought experiment is as disturbing as his 6 trillion Nazis example. But I have no idea what the number “6 trillion” really means, and am certainly not able to do moral calculus with numbers that large. But suppose we made both numbers smaller by a factor of one million, and removed it from the highly emotional Holocaust context. Now we are considering flu shots that merely protect against the inconvenience of 6 million one week-long cases of flu. Offsetting that is that 6 people (we don’t know who) will die from taking the shots. Are the shots worth it? That seems a much tougher call.
Before people accuse me of being heartless, let me emphasis that I do not disagree with Bryan’s moral intuition in the case he posed. I do question, however, whether those interesting thought experiments have anything useful to tell us about real world public policy debates. Indeed, I think the “message” that comes out of those examples, might well do more harm than good.
[BTW, I did not see the debate.]
Update: 4/16/09 I knew I would have to update this post. After reading the comments, especially the one by Bill, I want to make a few clarifying remarks. Bryan’s post was entitled “Are Grotesque Hypotheticals Cheap Shots?” My only goal was to address that question in the context of the debate over utilitarianism. Thus I took no position on the debate about liberty vs. efficiency, and indeed I am not certain whether I have a position on that issue. Another way of putting my point was that I was primarily interested in a methodological question; do hypothetical thought experiments about moral scenarios unlikely to occur in reality improve our moral reasoning? Most philosophers that I have known seem to answer yes. I say no. Alex Tabarrok reached a similar conclusion in this post from 2005. One of my arguments is the same as his argument.
The comments are still worth reading on other grounds. In particular, Bill makes some very effective arguments, for which I don’t have any easy answers. Thanks to TGGP for the Tabarrok link.
Update#2: In his reply (here) Bryan says I served up a softball, and I suppose he’s right. There are lots of ways to defend free trade. But I still think I have a better argument than it might appear at first glance. I should have constructed the hypothetical case more rigorously (I am still relatively new to this game) by having some assumptions like no impact on utility outside the examples mentioned in the case. I’m weak on trade theory—is there some sort of constant marginal cost assumption for China that would make their gains from trade infintesimal? And I do understand that if someone believes in freedom, they could defend free trade on those grounds as well. But what if the facts I provided were the only relevant costs and benefits? I would still come out for free trade on the cost/benefit criteria, despite the fact that if I used exactly the moral intuition that Bryan seemed to want me to use in the Nazi example, I might have gone the other way. It is also possible that I misinterpreted what Bryan intended by that example. I had assumed he wanted us to go with our gut instinct that applying efficiency in that case was wrong, even to a utilitarian. But he may have had other ways of reaching that conclusion (such as a revulsion to the loss of natural rights, which doesn’t even figure into the utilitarian calculus.) I’m still skeptical of these thought experiments for a number of reasons, but I agree there are ways around my specific case.
I also want to link (here) to Robin’s follow-up post. He asks a good question that I want to think about before answering. I am torn because I agree with Bryan that most people’s intuition about economics isn’t very good. On the other hand, like Robin I place a lot of weight on the “wisdom of crowds.” So for now I lean in Robin’s direction.
Tags:
15. April 2009 at 16:11
Scott,
People argument against the utilitarian approach is not right. They claim that under the utilitarian approach we would do things that are wrong like kill people because we value them less than what they value their life. That is the efficient outcome. It is true, but I see situation from another perspective. Suppose that person A values his life by $3,000,000. While the rest of us value killing him by $6,000,000. The utilitarian says then we have to kill him. But the thing is that we forget to pay A for killing him. So the rellevant moral problem is if it is ok for some markets to exists, markets for slavery, markets for life, markets children, markets for everything. So going to the case of slavery, in my opinion it was a bad thing not because slaves were abused but because they didn’t enter into the contract voluntarily. Since it is probably very hard to verify that a slavery contract is of mutual benefit to the concenting parts it was probably more efficient in a second best sense to ban slavery. This ofcourse generates some inefficiencies, like companies not investing in general human capital of their employees since the employee can then go and work for another company. Just like there is no market for slaves because it is illegal there are many other markets which are illegal like the market for live organ sales (kidneys, livers, hearts, lungs), the market for horse meat in California, the market for child pornography (and now for senior pornography in one state), or the market for hiring dwarves to be thrown in sporting events (I think this is done in France or England). Why are this markets illegal? Many of this “goods” have negative externalities on other people, like horse meat for human consumption, Californians are not worried about the horses since you can use horse meet in dog food but not in human food. No one will claim that the transaction between the horse meat seller and the buyer is not mutually beneficial. so the only possible explanation is that most californians (they voted on it) find the idea of other californians eating horse meat repugnant and it lowers their utility. Is the utility loss greater than the gain of eating horse meat to the horse meat eating californians? Well no cost benefit analysis was done and since we live in a democracy, the majority rules and the monitory has to suck it up. In the end horse meat eaters ended up being the “slaves”. Indentured servitude was made illegal because the transactions were not performed voluntarily (exploitation) and because again many people found the idea of somebody being a slave as “repugnant” (this term comes from Al Roth from Harvard). Now there is another reason why some markets are illegal which is really strange. People sometimes find that the good itself is not “repugnant” but the selling of the good for money is. This is the case for the market for organs. You can donate your kidney or part of your liver but you cannot sell. Now the market is the negative externality. Again, no cost benefit analysis is performed, no utilitarian rule is applied, but a simple majority rule. Now, if this is all about externalities then why doesn’t the Coase theorem apply? Well transaction costs are high because of the large number of people involved. I wrote too much, thanks for reading.
Alex.
15. April 2009 at 20:43
My moral intuition tells me that free trade is the right policy. Protection involves unwarranted interference with mutually beneficial exchanges between the people who produce barbie dolls at least cost and U.S. consumers of this product. Cost/benefit analyses would show that those who gain from a free-trade policy in the U.S. could generally over-compensate the losers.
You seem to be relying heavily on a distributional judgement that the “devestating effect” (utility loss) of the people in South Carolina who will lose their jobs would exceed the utility gain of the nice little girls whose parents will be able to afford to buy them barbie dolls. I don’t know enough about the circumstances to make that judgement. My moral intuition tells me that in most instances it is not appropriate to compensate rent-seekers who have have used their political muscle in the past to obtain the benefits of tariff protection at the expense of consumers.
15. April 2009 at 20:46
I’m sorry to mis-quote you, Scott. I mis-typed “devastating”.
15. April 2009 at 21:23
I was looking for one Alex Tabarrok post on the lives we accept losing in order to mine coal when I found this even more appropriate one where he supports Robin Hanson in biting the bullet and rejecting ethical/philosophical intuition.
15. April 2009 at 23:57
A couple of points;
Utility is an ordinal measure, so can not be compared across individuals. In the strictess sense, a transaction is `efficient` if the benefits outweigh the costs. On a practical basis, it`s just impossible to conclude a particular transaction is efficient unless both parties voluntarily agree to it while internalizing all benefits and costs. Under this definition, it`s impossible to conclude that any particular murder (ie, against the victim`s wishes) is efficient. I think what Bryan would argue is that once one comes up with a metric to weight different people`s utilities, it`s hard to consistently rule out senarios that one would otherwise object to.
Also, note that Bryan`s choice of measurement – `Liberty` – overcomes this hurdle. I think he is arguing that a transaction can be judged `good` if no affected party is forced into it. This rule seems to encompass the strict definition of efficiency.
15. April 2009 at 23:59
…ie, he would judge free trade to be `good`, because the buyer is free to stop buying from one farmer, and start buying from another.
16. April 2009 at 02:43
According to Caplan, Hanson’s version of “efficiency” is that any change should be implemented if those benefiting are willing to pay more than those who oppose it are willing to pay to avoid it. Caplan claims that no matter how horrifying the scenario he dreams up, Hanson will “accept” it. I think Caplan claims that Hanson favors implementing all Pareto improvements without compensation. If you beleive that property rights should be distributed according to where they would end up, (like the Coase theorm,) then this is where you are.
One element of this approach is that the distribution of income is considered irrelevant. The case of the rich cannibal and the orphan children points to that issue.
Aggregate happiness versions of utilitarianism don’t have that issue. It is implausible that the rich cannibal gains more happiness from eating people than the victims will lose. But it is possible that the rich person could pay more than the orphans and so, him being able to eat them is the “proper” allocation of property rights. If, for some reason, you argue that no, everyone should own themselves, then the rich cannibal must pay the orphans themselves, and this will be impossible. (Though we might come up with examples where the victims have bequest motives.)
The “too few Nazi” issue does apply to utilitarianism. On the one hand, there are distribution issues. Those who gain from killing the scapegoat may each gain just a little, but if there are enough of them, their willingness to pay could easily outstrip the wealth of anyone. But, this issue also relates to utilitarianism. You just need a big enough ratio of people made slightly happy by the suffering of others relative to the few who suffer greatly.
If we assume that we must get the permission of those who are to be made to suffer, that is, they have some property right in themselves, then this is not an issue. But according to Caplan, it is just that sort of self-ownership intuition that Hanson is trying to avoid.
I believe that rule utilitarianism seeks to finess this issue by asking whether a system of identifying scapegoats for suffering would in reality benefit most people. That is, hardly anyone would be willing to take the chance that this machinery would be turned against themselves or someone they love.
I think this is the answer to the free trade issue. A rule of always protecting people from competition will lead to universal impoverishment. If it was actually necessaary to compensate those actually harmed by competition, it could be done. (Where as, it isn’t clear that the trillions of Nazis could pay the millions of Jews to get them to die or that the rich cannibal could pay the children.)
If Caplan is defending an ethic of self-ownership, then the trade in dolls example isn’t a problem. Quite the contrary, it leads to the free trade result. The domestic doll producers have to get the doll consumers to agree to buy from them.
While Caplan’s account of Hanson’s version of efficiency allows for the free trade result as well as various muderous consequences Caplan uses as reductios, a self-ownership ethic and rule utilitarianism allow for free trade and not the murderous examples.
So, Caplan’s method doesn’t imply that free trade is wrong. But, it does suggest that it must be justified. It isn’t just about allowing poor American girls to enjoy playing with Barby dolls. Once we account for those gains, as well as the other goods that consumers of dolls enjoy because they pay less for the dolls, and the gains of those who sell the other goods, and the gains to the Chinese workers, and the gains to the Americans who sell to the Chinese workers, etc., etc., the moral outrage suggested by adding up costs and benefits can be overcome.
If Caplan’s account of Hanson’s views are correct, then I disagree with Hanson. I don’t think a rule of always checking willingness to pay and then adopting any change on that basis is correct. I don’t think that rule enhances aggregate human well being (I am a rule utiliatarian of a sort.) I do think rule utilitarian considerations lead to some kind of strong individual rights. On the other hand, I strongly suspect that the sorts of restrictions on individual rights implies by limited government are also justified by rule utilitarianism.
Hanson’s rule (according to Caplan) will often lead to the correct result. But not always, because it is the wrong rule.
16. April 2009 at 03:58
You suggest 1) utilitarianism or 2) moral intuitions, but there is a third alternative: moral principles, which might be counter-intuitive. One still has to justify the principles, of course, but you could do it in ways other than utilitarian calculus.
Another practical problem with utilitarianism is who gets to perform the calculations?
16. April 2009 at 04:52
After reading the comments (especially Bill’s) I added an update that clarifies my position.
Alex, Yes, I have read Al Roth and enjoyed his work. I was thinking of mentioning the organ transplant issue in my longer post on utilitarianism.
Winton, I also support free trade, even in the example I provided, and even assuming there are no extenuating circumstances. But this is based on reason; my moral intuition in that case would push me the other way. My reason tells me that my moral intuition is wrong.
TGGP, Thanks, I linked to it in the update.
Pedro, That’s a good point about Bryan’s defense of liberty, but as my update shows that wasn’t my real target. You are also correct that economists often assume ordinal utility, but that’s because we don’t know how to measure utility, and unlike non-economists we are to shy to impose our own values. Non-economists have no such inhibitions. If they are liberal utilitarians they might assume the value of a marginal dollar to a rich person is 1/10th that of a middle class person, and 1/100th that of a poor person (for purposes of public policy decisions.) Remember that the most common type of utilitarianism calls for maximizing aggregate utility, and that requires interpersonal comparisons.
Bill, Some of your points are addressed in the update. Your final paragraph about rule “utilitarianism” is a good one, and I agree. I see it as relating to my underlying pragmatism. This approach will likely lead to broad societal rules that are desirable, not the weird thought experiments dreamed up by philosophers. I do think however, that it points to a problem with utilitarianism. The term can be endless redefined so that it becomes almost tautological. Thus all utilitarians should favor adopting the version of utilitarianism that is best for society. This means that, in principle, debates among utilitarians about which version is best are often debates about means, not ends.
You are right that Hanson’s efficiency approach doesn’t account for distributional issues. But couldn’t Hanson argue that those problems could be solved with an optimal redistribution of income? And that once we had done so, Pareto optimality should rule? I don’t know if that is his position, but it would be a possible response from him.
I agree with almost everything you say in your reply. You are correct that the Nazi example does apply to utilitarianism in a way that some income inequality hypotheticals probably don’t. The analogy I would draw is to my bubble argument in the EMH paper. I said “yes, the 2006 housing market sure looks like an irrational bubble to me.” But then went on to argue that that intuition had no practical applications. I’d say the same about the 6 trillion Nazis. As soon as we scale it down to a level where it might be applied to real world societal problems, then I (and apparently Alex Tabarrok and Tyler Cowen) no longer find the intuition so persuasive. Although that’s also partly because the small number of victims who sacrifice themselves for the good of society are statistical lives, and are not singled out on grounds of ethnicity or religion.
16. April 2009 at 09:44
I clarify my views more here: http://www.overcomingbias.com/2009/04/efficient-economists-pledge.html , where you will also find a link to an audio of the debate.
Don’t know if it is original with you, but I really like the idea that we get our moral intuitions on what is bad from fictional cases where small gains are allowed to trump large losses.
16. April 2009 at 11:39
Kelo is the refutation of utilitarianism: People are incapable of wisely weighing one person’s utility against another, and if they try to, trouble ensues.
We must use other criteria, for if we use strictly utilitarian criteria, if we try to maximise total utility, we will fail. We will wind up with a society based on terror, mass murder, and famine.
16. April 2009 at 15:13
I don’t know if this point is too obvious to even make or not, but this post pretty clearly flows out of your previous post on political art. If a story involving events fairly close to reality (fictionalized depictions of historical events even) can’t serve as a persuasive critique of liberalism, then obviously sci-fi type hypotheticals (which are basically just mini stories) won’t be able to do so.
17. April 2009 at 03:45
Alex, Your question raises a number of issues that I hope to address Sunday. I have an open mind on the “principles vs. utilitarianism” debate. Regarding who decides–I think ultimately it must be the voters. Even if the proximate decision is made by government, or Constitutional provisions that limit government (as libertarians prefer) both of those must ultimately be decided by voters, who elect officials and enact Constitutions.
James, I would go back to Bill’s discussion of “rules utilitarianism.” Society’s total utility is probably higher if some things are decided using broad principles, rather than on a case by case basis. For me, the 1st amendment (which initially seems non-utilitarian) meets that criterion. I would also argue that the Constitutional prohibition against takings for private use (which the Court mistakenly refused to apply to Kelo) also meets that criterion. So I think that your view can be defended on utilitarian grounds.
Thanks Robin, I’m not sure if I was the first to use that argument.
blackadder, That’s an interesting point. I do see some similarities, but the cases are slightly different in the sense that an imaginary philosophical example attempts to eliminate any extraneous information. On the other hand, both are subject to the problem that much in economics (and I would add ethics) is very counter-intuitive. So that’s a good point.
19. April 2009 at 11:17
[…] shock our moral sensibilities. Yes, but only in principle. Since I addressed this issue in a post last week, I won’t say anything more […]