Monday, April 28, 2008

Does Studying Economics Make You Selfish?

There's been a lot of discussion in economics circles about how economics training makes people more selfish -- in particular, by teaching people "rational choice theory", the cartoon version of which portrays rationality as a matter of always acting in one's perceived (economic) self interest (for example, by defecting in prisoner's dilemma games and offering very little in ultimatum games). Accordingly, the economics literature contains a few much-cited studies that seem to show that economics students behave more selfishly than other students.

However, virtually all the experiments cited in support of this view are flawed in one of two ways. Either they test students on basically the same sorts of games discussed in economics classes, or they rely on self-report of selfishness. Relying on econ-class games makes generalizing the results very problematic. It's no surprise that after a semester of being told by your professor that defecting (basically, ratting on your accomplice to get less prison time) would be the rational thing to do in a prisoner's dilemma game, when that same professor or one of his colleagues gives you a pencil-and-paper version of the prisoner's dilemma, you're more likely to say you'd defect than you would otherwise have been (even with small real stakes). What relationship this has to actually screwing over acquaintances is another question.

Likewise, relying on self-report of selfishness is problematic for all the reasons self-report is usually problematic in the domain of morality, and in this case there's an obvious additional confound: People exposed to rational choice theory might feel less embarrassed to confess their selfish behavior (since it is, after all, rational according to the theory), and so might show up as more selfish on self-report measures even if they actually behave the same as everyone else.

I've found so far only three real-world studies of the relationship between economics training and selfishness, and none suggest that economics training increases selfishness.

(1.) Though I find their study too problematic to rely much on, Yezer et al. (1996) found that envelopes containing money were more likely to be forwarded with the money still in them if they were dropped in economics classes than other classes.

(2.) Frey and Meier (2003) found that economics majors at University of Zurich were less likely than other majors to opt to give to student charities when registering for classes, but that effect held starting with the very first semester (before any exposure to rational choice theory), and the ratio of economics majors to non-economics majors donating remained about the same over time (all groups declined a bit as their education proceeded).

(3.) Studying professional economists, Laband and Beil (1999) found a majority to pay the highest level of dues to the American Economic Association (dues prorated on self-reported income), though they could without detection or punishment have reported lower income and so paid less. Through an analysis of proportion paying dues in each income category vs. proportion in the profession making income in those categories they found similar rates of cheating in self-reported income among sociologists and political scientists.

I see these findings as the flip side of what I've been finding with ethicists: Just as ethical training doesn't seem to increase rates of actual moral behavior much, if at all, so also being bathed in rational choice theory (if, indeed, this is what economics students are mostly taught) doesn't seem to induce real-world selfishness.


  1. I like very much these ongoing studies on the relationship between acquired expertise and skills and the subsequent moral outcome during the exercise of the "professions". It is a truly experimental philosophy case where philosophical theory is put into the real world.
    I would like to mention another kind of "profession": seminarists or theologicians.

    The well known Darley and Batson´s Good samaritan experiment: Darley, J. M., and Batson, C.D., "From Jerusalem to Jericho": A study of Situational and Dispositional Variables in Helping Behavior". JPSP, 1973, 27, 100-108.
    where those investigators found no effect with being a seminarist or theologician and being a "good samaritan".
    With these classical experiemnts and yours is probably not related any personality trait or variable with behaviour (moral) outcome.

    But Why? personality traits are acquire thoughts and habits of thought that chanels our emotions and behaviour, how is possible that someone who "trains" his personality then act in oppossition to it.

  2. Thanks for the comment, Anibal. The original subjects in Darley and Batson 1973 were all Seminary students, so no inference is possible about whether non-seminary students would be more or less likely to have stopped to help. (They did find that those testing higher on a "religiosity" scale were no more likely to help, however.)

    I agree with the tenor of your final question -- personality traits must have some important relationship with moral behavior. But it's amazing how tenuous and elusive that relationship seems to be!

  3. Eric:

    I believe the root issue here is the classic "knowing-doing gap." No matter what the study or profession, we are taught to "talk smart," but not taught to put our knowledge into action.

    Being a college professor, I am sure you witness this behavior everyday, perhaps in your own classroom. Have you seen students get better grades for "participating" in class discussion? Displaying our knowledge does not display our ability to apply our knowledge and our students are "rewarded" for talking smart with little thought or reward in place for actually acting on that knowledge.

    There is rarely any application in professional studies or in actual professional environments of aligning our knowing and our doing together.

    The "selfishness" you speak of results from the self-given idea that ethics apply to others but not to ourselves. Once again, I believe this selfishness is a result of being taught to "talk smart" but not act accordingly.

    The "knowing-doing" gap must be closed!

    If you have not read "The Smart Talk Trap" by Harvard Business, I highly recommend it:

  4. Thanks for the comment, I very much enjoyed it. I would also like to point your attention to some work done in economics with regards to charitable giving.

    Also, I think there is a subtle difference between rational thought and selfishness - something, sadly, in which there is not enough time to fully highlight in undergraduate courses. Without going on a lengthy ramble discussing this, I will point out that the often used example, Prisoner's Dilemma, is a very particular situation. It is a static scenario with no threat of retaliation. Moreover, the payoffs are only characterized in prison terms...I digress, its goal is to teach students to be aware of all possible states of the world when making a decision. Again, this can be a difficult concept to fully characterize to undergraduates and a student could walk away with the impression that economics teaches them to screw over their friends.

  5. Thanks, Financial and anonymous!

    Financial: I agree that part of the issue here is the gap between talk or writing (what is rewarded in school) and action. It's bad enough in science, from the stories I hear, let alone philosophy! I can't agree more that "smart talk" is rewarded in philosophy -- and I also believe that confusing or impenetrable smart talk is often rewarded at the professorial level. This, I think, partly explains Hegel and all those incomprehensible APA talks.

    Anonymous: I tried to follow your link, but it didn't seem to work -- can you try another one? The studies I've seen of economists and charity are based on self-report, which I'm suspicious of for reasons I've explained. Frey and Meier is an exception (and even better has a time course that can be related to economics training); are there other exceptions too that I haven't noticed yet?

    I agree that rational choice theory doesn't have to be as horribly selfish as it sometimes comes across, especially in simple presentations. For example, the well-being of others may be an important good to you, and in public settings and for repeated games there's reputation, etc....

  6. Try this

    then go to the section on Altruism and Neuroeconomics. I am not Bill Harbaugh, by the way, this is my first blog comment, so I didn't have a name.

  7. Thanks for the link, anon. That looks like an interesting article!

  8. I've heard rumors about these studies for years, but I've just recently been moved to take a closer look. I still worry that the preponderance of even cartoonish rational choice studies precludes the possibility of other approaches, ones I'd find more morally attractive, to be given air. Even if studying economics doesn't make one selfish, I wonder what the diffuse, superficial study of economics has done to our national culture and conventional wisdom. Studying Christianity doesn't make one more Christ-like, but I imagine at least Weber would say that there are some externalities that accompany the wide-spread study.

  9. I agree, Snowden, that it's unlikely to be entirely without effect. But what exactly the effects are, is hard to tease out.