The benevolence of self-interest

      Critics of economics often accuse the discipline of viewing people as mere
      optimising machines, as ethical nonentities. The charge would be serious if it
      were true—but it is in fact false
 

      HOW can economists expect to be taken seriously, non-economists are given to
      complain, when their model of man is so patently inadequate? Mainstream economics
      assumes that people are driven by the rational pursuit of self-interest. But, as
      everybody knows, people are not rational and they often act selflessly. Where in this
      view of man as desiccated calculating-machine is there recognition of the centrality of
      love, duty and self-sacrifice in human conduct? What use is a purported science of
      social behaviour that is blind to the necessary conditions for social behaviour?

      These questions would be telling if “rational” and “self-interest” meant what these
      critics take them to mean. But they do not. In mainstream economics, to say that
      people are rational is not to assume that they never make mistakes, as critics usually
      suppose. It is merely to say that they do not make systematic mistakes—ie, that they
      do not keep making the same mistake over and over again. And when economists talk
      of self-interest, they are referring not just to satisfaction of material wants, but to a
      broader idea of “preferences” that can easily encompass, among other things, the
      welfare of others.

      Even when the terms are properly understood, “rational pursuit of self-interest” is a
      simplifying assumption. But the right question is whether this simplification is fruitful, or
      so gross that it hides what needs to be examined. Human behaviour is far too
      complicated to be analysed—to yield patterns and suggest generalisations—without
      employing some such simplification. And in nearly every branch of economics,
      rationality has proved a useful one.

      There are some exceptions: much effort has recently gone into examining non-rational
      or nearly rational behaviour in special contexts, often with interesting results. It is right
      to be open-minded about such things. But critics of economics, if they believe that any
      kind of social analysis is possible, had better say what other simplifying assumption
      they would rather use. Unsurprisingly, there is no plausible candidate: on the whole,
      people do learn from their mistakes.

      Turning from means to ends, what about self-interest? Here the issues are subtler. If
      economics supposed, at one extreme, that people seek only to maximise their material
      consumption, then it would be plain wrong, and that would be that. If, at the other
      extreme, it assumed that people seek to satisfy their preferences (or some such
      formula), then it would be true merely by virtue of the meaning of the words—and it
      would not tell you anything. The assumption built into mainstream economics is much
      closer to the second of these than the first. Anti-economists who find it absurd would
      be nearer the mark if they called it a statement of the obvious.

      However, the assumption of self-interest is not entirely tautological. Many kinds of
      apparently selfless behaviour may in fact be self-interested in the way economics
      proposes—but not all.

      Into the first category, selfless behaviour that is not, fall acts that invite or assume
      reciprocity. These are the everyday transactions (if you will forgive the term) of living in
      society. People show consideration for others in the hope or expectation that the
      favour will be returned. Behaviour that establishes a reputation for honesty, or that
      signals a willingness to enter into commitments, is also, as a rule, self-interested in this
      sense. That makes it no less conducive to a flourishing society, no less to be praised
      and encouraged. Fortunately, it is self-interest, not love, that holds society together.

      Into the second category, the realm of the purely selfless, fall acts such as sacrificing
      your life for a stranger—or, less dramatically, leaving a big tip at a restaurant you will
      not be visiting again. If acts that were both costly and purely selfless were common,
      outside the family or other close relationships, economics would be in trouble.
      Analysing a labour market in which workers demanded more work for lower wages,
      or employers wanted to lose as much money as possible, would present some
      difficulties.

      Sad to report, society at large might be in trouble too: universal altruism would, it
      seems, upset the basis for social co-operation. (“I’ll do you a great favour, but I insist
      that you don’t do me one in return.” “Sorry, no. I’ll do you a great favour, but only if
      you don’t return the compliment.”) In any event, acts of heroism are rare. Like
      rationality, self-interest (even when broadly defined) fails to capture some aspects of
      social behaviour, but not so many as to render models based on the notion useless.
      Again one must ask what other simplifying assumption would serve better. Again, none
      has been suggested.

      The unrepentant anti-economist might retort: better no analysis than so gloomy a
      science. This is doubtless a matter of taste and temperament. But is it really so
      gloomy? When Adam Smith pointed out that, if people want dinner, they look not to
      the benevolence of the butcher, brewer or baker, but to their regard for their own
      interest, his aim was not to portray social interaction as mean and narrow. Rather it
      was to draw attention to the extraordinary and improbable power of self-interest: this
      stunted, inward-looking trait is transformed, through spontaneous social co-operation,
      into a force for the common good.

      Smith regarded this as almost miraculous. So it is. The main task of economics has
      been to understand this astonishing process. And by and large, thanks to its simplifying
      assumptions, it has succeeded. That’s not so dismal, is it?