Rrpnnted for private circuLtkm from Economic Inquuy Volume XVII, Number 4 OCTOBER 1979 PRINTED IN USA AN EVOLUTIONARY MODEL OF TASTE FOR RISK PAUL H . RUBIN and CHRIS W. PAUL II' Economists take tastes as given. However, tastes must be derived from biological models of evolutionary survival; we exhibit those tastes which served to make our ancestors survive. In particular, economists have no theory which explains ob- served behavior towards risk; rather, we take behavior as a datum. In this paper we present a model which explains risk seeking by adolescents and risk aversion by mature males as the result of an eoolutionary mechanism. While textbooks indicate that changes in tastes are a potential source of changes in demand, empirical and theoretical work ignores tastes. As a research methodology, this strategy has been extremely productive - for example, the work of Becker and his followers on areas of behavior as diverse as crime and marriage has given fruitful results without con- sidering tastes at all (see Stigler and Becker (1977)). The presumptionhas been that tastes are constant (over space and time) and that therefore changes in behavior can be explained in terms of changes in relative prices. The other justification for ignoring changes in tastes is that we have no theory as to when and how tastes change, and thus it is impos- sible to make falsifiable statements on such changes. There exists, however, another body of literature which may have implications about tastes. This is the literature of "Sociobiology" (Wilson (1975)) - the emerging discipline dealing with biological deter- minants of social behavior. (See also Becker (1 976), Hirshleifer( l977), Rubin, ut. al. (l979), Tullock (19771, and Fredlund (1976) for contri- butions by economists to this literature, and Ghiselin (1974), for a con- tribution l)y a biologist using economic arguments.) Viewing human t~ehavior in terms :)f evolution indicates that tastes may be the result of selection pressure - that is, the drives and goals of humans may have been selected for, in the sense that our ancestors who had such drives and desires in fact survived to become our ancestors, while their relatives with less fit (in the biological sense) goals did not survive or reproduce. (Coase (1978) argues that sociobiology should enable us to determine the ele- ments and form of the utility function.) In this sense, it is useful to view " natural selection as maximizing, in an as if" selise, for this method- ology, if used correctly, would give valid predictions. In using biological arguments, it is important to specify the level of which selection operates - that is, does selection operate at the level of *Thc University of Georgia. The authors would like to thank two anonymous referees for helpful commen ts. Economic lncluiry VoI. X\'ll. Oct. 1979