Economics & Sociobiology Copyright (c) 1994 by Nick Szabo Sociobiology looks a lot like economics. In particular, they both use game theory. Cf. the work of John Maynard Smith, Robert Trivers, William Hamilton, et. al.; and the popularization by Richard Dawkins, _The Selfish Gene_. There is a fundamental difference from economics, in that the "agents of self interest" are not individuals, or groups of people, or cultural information. The strategies are rather embodied in developed neural and hormonal structures, which are encoded by cooperative combinations of alleles (alternative genes). Each gene codes for a protein, and thus different alleles code for different forms of that protein, and different combinations of alleles for different combinations of alternative forms of the protein. For game theoretic purposes this genetic and developmental detail is abstracted to a transmission of genetic information that encodes a behavioral strategy. Smith demonstrated how the process of genetic selection gives rise to strategies that approximate strategies arrived at by "rational self interest". Where economic agents learn these strategies, either via tradition or the process of playing the economic games, genes only "learn" via the evolutionary process of variation and selection. Where economic agents use thought, genes create metabolic and neuronal structures that are constrained to perform the game strategies, called evolutionary stable strategies. ESS's are a variation of Nash equilibria, computationally bounded by the natural selection process (for arriving at the strategy), and by embryological development and metabolic constraints (for implementing the strategy as behavior). Alas, the translation of anthropomorphic terms from economics to biology, like "rational self interest" to describe genes (or in Dawkin's more succinct metaphor, "selfish genes"), has brought a large amount of misunderstanding about the ethical aims of the enterprise. As they have in economics, for that matter. But there's less excuse for confusion in biology, as the algorithmic process of evolution, and its mathematial relationship to game theory, is much better characterized than the notion of "rational self interest" in economics. Going further, I'd argue that what is mathematically "rational" at the level of genetic evolutionary strategy, what sociobiologists call the "ultimate" level, is most likely to correspond to what we consider emotional and irrational on the behavioral or "proximate" level. Sociobiology is in some ways a postmodern Freudianism, although it is far mar rigorous and comes to very different conclusions than Freud. Interestingly, I've found the conclusions of human sociobiology in fascinatingly close agreement with the wisdom of blues music; evolution explains blues lyrics better than any other theory I know of (although, alas, no theory explains blues music as well as real life!) Some have also found strong resonances between sociobiology and Calvinism (eg, our genetic endowment as "original sin"). Creationists and evolutionists in agreement after all! Perhaps the most important result of sociobiology for economics is that, in terms of our genetically endowed constraints, humans in fact stray quite far from the rational self interest ideal at the level of the individual. An obvious example is kin altruism: most obviously the mother-child bond, but there are many other forms. Some form of Axelrod's "tit for tat" reciprocal altruism may also be genetically endowed, as behavioral strategies and emotional biases that appear via neuronal and hormonal development, biasing the learning process. Instead of dumping "rational self interest", economists might instead interpret genetic endowments as part of our particular utility functions; in which case sociobiology might be incorporated into economics as a science of utility. There may, for example, be strong correlations between one's time preference and one's "genetic state" (states of life that are important in terms of genetic evolution, such as age, male vs. female, coupled vs. single, gay vs. straight, number of siblings, children and other close kin, marginal fertility expectancy, etc.) In any case sociobiology allows economics to replace the current tautological relationship between "rational self interest" and "utility" with a scientific model of genetically endowed strategies, which bias the process of learning strategies from economic experience and the cultural endowment of tradition. Large deviations from the rational self interest model are confirmed by experimental game theory; observed results diverge sharply from even computationally trivial Nash equilibria. For a good lay description of these results see Thaler, _The Winner's Curse_. (As a bonus, many tips on how to take advantages of deviations from rational self interest for fun & profit! George Soros, perhaps our era's most successful financial speculator, comes to similar conclusions in his _Financial Alchemy_). To avoid the inevitable flame war on nature vs. nurture (fat chance!), any reasonable person on either side realizes that we are irrevocably a mixture of both, and any social scientists who even comes close to deserving the title "scientist" is open to explanations of either kind, or (probably the most typical case) explanations where innate and learned behaviors both come into play. I think reasonable sociobiolgists also agree that nature is not destiny in humans; it is just sometimes a god-awful annoying pain in the butt. :-) In terms of economic agents, nature is a computational constraint: it makes some solutions computationally more feasible than others; it makes some equilibria more likely than others; this can make a big difference in a world of bounded reasoning.