Reviewed by Eric W. Noreen University of Washington
In this book, Robert H. Frank, who is a professor of economics at Cornell University, persuasively argues that passions (or emotions) play a strategically important role in human affairs. Paradoxically, our passions, by leading us to forgo tactical mate-rial advantages, may actually result in strategic material advan-tages. This notion is strange enough that I think it is best to dis-cuss it within the context of an example drawn from behavioral laboratory studies cited in the book. The experiments were con-ducted using pairs of subjects who did not know one another and each subject participated in only one trial. Subject 1 was given a sum of money and was asked to propose a division between himself and subject 2. Subject 2 then had the option to accept or refuse the division. If the offer was refused, neither subject got any money at all. The “rational” strategy for subject 2 is to accept any division that leaves him with some cash since the alternative is zero. Knowing this, the “rational” strategy for subject 1 is to propose a division in which he gets nearly everything. In fact, in about a quarter of the trials, subject 1 proposed a 50-50 division and the average percentage demand by subject 1 was 67% — a far cry from the 99+% predicted by the theory of rational economic man. Moreover, in about 22% of the trials the divisions proposed by subject 1 were rejected by subject 2 — even though in all cases subject 2 would have been better off accepting the offer. And, quite clearly, there was a relationship between the “unfairness” of the proposed division and the tendency for the division to be rejected.
This single experiment, which is only one of many cited in the book, raises all sorts of interesting issues. First, the results are clearly at odds with what we would ordinarily think of as rational behavior. These were all one-shot games between subjects who did not know each other, so there were no strategic advantages, within the context of the experiment, for subject 1 to be magnanimous or for subject 2 to reject an offer — however small. And yet such behavior was regularly observed. Emotions are involved in this experiment and the fundamental point made by Frank is that such emotions are useful — not in every situation necessarily, but in enough situations that they confer advantages on those who have them.
Why did subject 1 so often propose a 50-50 split instead of a 99-1 split as “rational” theory would seem to predict? A rationalist might respond that subject 1 was afraid that subject 2 would reject a lop-sided division; but of course this could only happen if subject 2 were himself irrational. Indeed, the fear that subject 2 is irrational may motivate subject 1 to give subject 2 a greater share! However, another force is at work that also mitigates against the predicted 99-1 division. Namely, most of us feel bad if we take unfair advantage of others. Now what possible advantage can such guilt feelings have?
An indirect (and unintended) lesson of agency theory is that opportunistic behavior generates enormous deadweight losses. Markets fail, sub-optimal decisions are made, monitoring costs are incurred, etc. A sense of guilt can act to counter opportunistic behavior and thereby reduce deadweight losses due to agency costs. Moreover, if there are statistically reliable signs concerning who feels guilt and who does not (such as certain facial expressions), those who feel guilt will be sought after in ventures that require mutual trust. The capacity to feel shame and guilt therefore conveys important material advantages; in the economists’ jargon, it enlarges an individual’s opportunity set.
An important insight is that, by their very nature, emotions are involuntary and it is this aspect that solves a variety of prob-lems in economics and relationships. If I could turn my feelings of guilt off and on at will, you would not trust me to feel shame if I were to cheat you. Hence you would be reluctant to enter into any venture with me that presents me with opportunities for exploiting you. Unfortunately, since emotions are involuntary, they may be activated at times that are inappropriate as well as when they are in our long-run interests; on balance, however, Frank argues that they are useful. Due to natural selection, the very existence of emotions provides prima facia evidence of their usefulness.
Returning to the experiment, why is it in subject 2’s inter-ests to reject a division that would leave him with something — however small. Clearly, it is not rational for subject 2 to reject even a token payment — and yet he does. The theory of rational economic man simply does not correctly predict behavior in this situation. Frank argues that subject 2 refuses the offer out of anger over an “unfair” division. Of what possible use is such an emotion of anger?
Strategically, an individual can gain in the long run if he can convince others (via a reputation or involuntary signals) that he will act irrationally and refuse advantageous, but “unfair”, offers. The adversary of such an irrational opponent is forced into making a “fair” offer. Once again, this emotion has to be involuntary to be credible. In the context of the experiment, the emotion triggers a response that is clearly not in the interests of the individual. However, if the individual had the ability to turn off the emotion in situations in which it would be to his immediate material advantage to do so, the emotion would not have a deterrent effect. The emotion must have the power to override the “rational” response in order to be strategically useful.
A last example from the book brings this point home. In a world populated by rational men as modeled by economists, Mutually Assured Destruction would not be a credible nuclear defense. The victim of a first strike has nothing to gain by retaliating. Indeed, retaliation would likely result in the total destruction of all life. Hence, no rational being would retaliate. Knowing this, a rational being would strike first. Thankfully, our leaders are irrational and can be trusted to become angry and push the button in retaliation. The emotion of anger makes the threat credible and thus prevents the first strike.
In this review, I have just scratched the surface; there are many fascinating insights in the book backed up by impressive empirical evidence. Passions Within Reason should be required reading for all social scientists — particularly those of us who have been influenced by the rational economic man paradigm.
