Book review for PHILOSOPHY OF THE SOCIAL SCIENCES, Vol. 37 No. 4, 2007
Gintis, Herbert, Samuel Bowles, Robert Boyd, and Ernst Fehr, eds. Moral Sentiments and Material Interests. Cambridge MA: MIT Press, 2005. Pp xii + 404. US$ 50.
The main postulate of standard economic theory states that every economic agent attempts to maximize his expected utilities regardless of other people’s interests. This is what nearly all economists and their imitators have been saying for 230 years: that we are all basically selfish bastards. This must be true, since it has gone unchallenged in thousands of economics textbooks. And also because it is essential to any apology of the so-called free market, i.e., the unrestrained capitalism painted in books and political manifestoes.
In fact, the postulate is so prestigious that it has spilt over from economics into all of the social sciences under the name of ‘rational choice theory’: economic theories of marriage and divorce, child rearing and education, crime and crime prevention, voting and religious observance, and so on. It has even been adopted by popularizers of genetics (Dawkins’s “selfish gene”) and by the sociobiologists and evolutionary psychologists who are puzzled by altruistic behavior. They all tell us that cooperation, far from being beneficial to both individual and group, is costly and therefore threatens the cooperator’s Darwinian fitness.
If this is true, the occurrence of cooperation must be due to some dark genetic mechanism, such as inclusive fitness, according to which you will do something for your offspring–but not for your spouse, best friend, favorite political prisoner, or partner in crime, because you are not biologically related to them. The facts that criminal gangs and power elites, both of which are rather cohesive, are composed mostly of biologically unrelated individuals, should not be allowed to besmirch William Hamilton’s beautiful formulas. (Incidentally, most of the contributors to this book pay lip service to group selection theory.) If you are a biological reductionist, biological relatedness is far more important in social matters than social relatedness and commonality of interests.
Nor should it bother anyone learning that nobody knows enough about behavioral genetics to explain differences in social attitudes in terms of genic differences. When it comes to human nature, it is enough to decree either that “Genome is destiny” or that “Experience is destiny”. Either slogan is likely to find millions of followers, because in fact both nature and nurture determine behavior, though so far we hardly know how to combine them other than waiving our hands, the more vigorously the less we know.
The same holds of course for the selfishness principle, which we see confirmed every day, and which suffices to build any number of simple mathematical theories of human behavior. Maybe utility maximization is written in our genes? Perhaps there is a capitalism gene along with a God gene? And what if the expression of one of them were to either stimulate or inhibit the expression of the other? Think of all the formulas that could be invented and all the papers that could be published on these grave matters before being exposed by any pesky laboratory scientists.
To better appreciate the intuitiveness, simplicity and beauty of rational choice theories, let us manufacture a simple example: an economic theory of cheating at exams. A single obvious postulate suffices: that it is worth cheating if the expected gain G is greater than the expected risk R. In turn, we may set R equal to the penalty P multiplied by the probability p of getting caught. (We are assuming, as is usual in rational-choice theory, that the cosmos is a casino, so that every event can be assigned a probability.) In self-explanatory symbols, G > pP.
For example, let G be the number of course credits at stake, and P the loss of the same number of credits for flunking the course because of cheating. Since G = P, for cheating to be worthwhile the probability p of being caught has got to be less than 1. And since even such a high value as p = 0.99 meets that condition, the theory predicts that cheating is nearly always desirable. So, two birds are winged with a single shot: decency and the real goal of education, which the ignorant of rational choice theory is naive enough to believe to be learning. All behavior types and all kinds of knavery are collapsed onto one, and morals can be safely ignored. What better proof of the advantages of economic theories of social behavior?
Rational choice theories are not confined to explaining human behavior. The book under review makes it plain that most social policies presuppose the selfishness postulate. Indeed, they assume that the individual cannot be trusted to make spontaneously his fair contribution to society: We would all be basically “defectors” or “free riders”, hence in need of being forced to contribute to the res publica. This was the central message of Mancur Olson’s influential Logic of Collective Action (1965).
Seen in this pessimistic light, the very existence of voluntary and self-governing associations, such as cooperatives, mutual aid societies, PTAs, car pools, private anti-crime associations, Masonic lodges, cultural associations, learned societies, the Red Cross, and Amnesty International, could only be explained either by widespread hypocrisy or by the existence of isolated pockets of fools who don’t know what’s good for them. Actually not even families should exist, since every member of a family old enough to coordinate his or her movements is expected to be a cooperator, however reluctant, rather than a shirker.
However, let us for a moment rise above those annoying facts and suppose, just for the sake of argument, that the selfishness postulate were tested and found to be false. What would happen? If falsity were enough to drop a theory, the thousands of “rational” choice theories would quickly be consigned to the dustbin of intellectual history. But of course this is not how things happen in the fields where ideology trumps objectivity: there, mathematical morticians know how to make dead theories look alive. As Milton Friedman wrote in the 100th volume of The Economic Journal , contemporary economic theory is only “old wine in new bottles”. Wine or snake oil?
Still, suppose that a bunch of economists were to try and put the selfishness postulate to the empirical test, on the naive methodological ground that, since even the most unrealistic economic theories claim to refer to the real world, they should stand or fall according as they agree or disagree with the relevant empirical evidence. As a matter of fact a few such miscreants have been around since the 1970s: they call themselves ‘experimental economists’.
The first generation of this new species of economists, in particular Herbert Simon and James March, as well as Daniel Kahneman and Amos Tversky, focused on the rationality aspect of the postulate. They found that real people do not calculate their decisions as coolly and accurately as the theory predicts. For instance, far from waiting for the most promising business deal to appear, flesh and blood businessmen grab the first opportunity likely to defray their overhead costs. And the subjects become irrationally attached to their possessions even after they have outlived their usefulness: they hoard them instead of selling them at a profit. It is conceivable that experimental economists will eventually discover that people may become so attached to other persons, that they may wish to live together with them. If this came to happen, then even economists and their imitators might start using such unacademic words as ‘friendship’ and ‘love’.
However, let us get serious. Over the last decade a second generation of experimental economists has stolen the limelight. Curiously, some of the most innovative among them live in that model of conservative countries, namely Switzerland, and their family names start with F: Ernst Fehr, Armin Falk, Urs Fischbacher, and Bruno Frey. The first three F’s work at the Institute for Empirical Research in Economics, University of Zürich, whereas Frey is an emeritus professor of the famous Zürich ETH (where he was my chairman while I taught there in 1973).
The gist of these findings is that, while there certainly are some “rational egoists” among us, such as the proverbial Enron tycoons and what Geoge Soros calls “market fundamentalists”, they constitute only about one-third of the total population. (In a famous experiment, Richard Thaler and colleagues showed nearly two decades ago that economics students, unlike humanities students, tend to behave the way standard economics stipulates.)
The majority of experimental subjects turned out to be reciprocators, in particular what the contributors to the book under review call “strong reciprocators”. These have “a propensity to cooperate and share with others similarly disposed, even at personal cost, and a willigness to punish those who violate cooperative and other social norms, even when punishing is personally costly and cannot be expected to entail net personal gains in the future” (C. M. Fong, S. Bowles, and H. Gintis, p. 282).
Notice the difference between strong reciprocity and reciprocal altruism, which is returning good for good and bad for bad. The reciprocal altruist banks on future encounters, whereas the strong reciprocator behaves this way even in single-shot exchanges: he has moral sentiments in addition to material interests.
Strong reciprocity explains much economic reality that “rational egoism” does not. For instance, wage stickiness– the fact that wages hardly decrease during periods of depression–is explained by the desire of employers to keep up high employee morale, which is not just work satisfaction but also loyalty and eagerness to do high-quality work. Reciprocity also explains why people respond better to inner motivation (e.g., the joy of learning) than to external reward (e.g., grades). In particular, it explains something that R. M. Titmuss had found several decades ago: that people donate their blood gladly as long as blood is not marketed. Apparently, the market stunts moral sensitivity and civic virtue. It turns citizens into mercenaries, friends into consumers, patients into customers, and so on.
According to standard economics, people are motivated only by external interventions, in particular extrinsic rewards, such as earnings and bonuses. However, three decades ago psychologists found the “hidden cost of reward”, namely, that it tends to depress intrinsic motivation. And more recently experimental economists, in particular Bruno Frey, have found that the effect of external interventions depends upon the way the subject perceives them. They “crowd out” (weaken) intrinsic motivation if perceived to be controlling; but they “crowd in” (exalt) intrinsic motivation if perceived as supporting (E. Ostrom, p. 260). The corrupt public servant won’t shuffle papers unless flattered, bribed, or threatened, whereas the starving concert pianist will be sustained by his audience’s applause.
In general, though, the main social effect of external rewards is that they tend to forestall self-regulation, which is just as essential to the democratic governance of any social system as to the conduct of the autonomous and responsible moral agent. In other words, when external reward is the only motivation, and moreover when it is beyond the subject’s control, it generates dependency, passivity, and conformism. Treat people like lazy bums and they will tend to behave as such. Ditto with punishment: “severe penalities crowd out and mask the disposition of individuals to community self-policing, making it all the more necessary to employ severe penalties” (D. M. Kahan p. 359). Treat people like knaves and they will tend to behave as such. So, the dominant systems of reward and punishment, as well as of social control, have the perverse consequence that they erode positive liberty, the freedom to do, often in the name of freedom.
An impact of these experimental results on policy-making is that people feel civic duties when they are consulted and involved rather than being the passive objects of bureaucratic decrees. For instance, people will protest if a nuclear waste repository is placed in their neighborhood (NIMBY) even if, or rather particularly if, they are offered monetary compensation. But they won’t complain if shown that other communities accept, without compensation, similar public burdens just out of civic duty. Likewise, people pay their taxes as long as they believe that everyone else pays them; but they tend to cheat on their tax returns when they know (as every Italian tax-payer does) that most rich people get away without paying taxes.
So, now we have empirical evidence for thinking that we are not nearly as bad as economists have painted us. Kropotkin was far closer to the truth about human nature than all the standard economists put together: There is cooperation along with competition, partly because people have the very same moral sentiments (sympathy and empathy) that Adam Smith had noted in his 1759 book on the subject, but forgot in his 1776 foundational work on the wealth of nations. And also partly because the very idea of a society of fully egoist individuals is just as much of an oxymoron as that of an anarchist state. Cooperation is needed to attain any social objectives, such as security and public health.
The well-read reader will point out that many scholars, among them Aristotle, Ibn Khaldûn and Spinoza, knew that most people are not as selfish as Hobbes, Hume, Smith, and their followers tried to make us believe. They will also note that a few economists, as well as much fewer philosophers of economics, have criticized the postulates of “rational” choice theory. Surely, as Henri Poincaré said a century ago, criticism is the salt of science. Still, it is the salt, not the meat and potatoes. Now we realize that the meat and potatoes that the economists have been serving us were spoilt. Consequently, economists and policy scientists will have to restart form scratch.
But replacing the”rational egoist” of standard economic theory with the “strong reciprocator” of experimental economics will take far more than great ingenuity: it will also require a lot of intellectual courage, because the endeavor will be seen for what it is, namely a challenge to the ruling ideology. And economists have been trained, for more than two centuries, to spread the gospel of economic liberalism (the superiority of textbook capitalism) rather than to realign the economic order with human nature as revealed by experiment instead of dogma.
Gintis, Herbert, Samuel Bowles, Robert Boyd, and Ernst Fehr, eds. Moral Sentiments and Material Interests . Cambridge MA: MIT Press, 2005. Pp xii + 404. US$ 50.