Where do preferences come from? What determines their properties? Though traditionally reluctant to ask such questions, economists have recently turned to evolutionary models for answers. We focus on intertemporal preferences here, arising out of the evolutionary implications of different reproductive strategies or life histories. An agent’s life history specifies the agent’s number and timing (and, in a richer model, such aspects as quality) of offspring. Evolution will select the life history that maximizes the growth rate of the associated group of individuals. We begin with the simplest possible biological life history—that of a semelparous agent that, if it survives a fixed number of years, reproduces and then dies. We can solve for the growth rate explicitly in this case. The evolutionary criterion for success entails hyperbolic time discounting of the log of the number of offspring produced. However, the rate of time preference is a function of age, not of time relative to the present, and hence there are no preference reversals in the sense of behavioral economics. At the same time, the optimal

Higher ability workers benefit more from bigger cities while housing costs there are higher for everyone, and yet there is little sorting on ability. A possible explanation is that young individuals have an imperfect assessment of their ability, and, when they learn about it, early decisions have had a lasting impact and reduce their incentives to move. We formalize this idea through an overlapping generations model of urban sorting by workers with heterogenous ability and self-confidence, with the latter defined as individuals ’ assessment of their own ability. We then test the location patterns predicted by the model over the life cycle on panel data from the National Longitudinal Survey of Youth 1979. We find that the city-size choices of individuals at different stages vary with ability and self-confidence in a way that closely matches our theoretical predictions.

research assistance. Hastings gratefully acknowledges financial support from the National Institute

helpful comments, to Jesse Shapiro and Andrei Goureev for excellent research assistance, and to the National

Oster, Alan Sorensen, Jesse Shapiro, and especially J.P. Dube for helpful comments and assistance with the

We thank Jesse Shapiro for early conversations which encouraged us to write this paper, as well as

617-627-3712. I thank Michael Gnade and Charlotte Taylor for research assistance. I thank seminar participants at Rice and Aimee Chin, Sam Peltzman, Jesse Shapiro and Cass Sunstein for valuable discussions. 2

We would like to thank Editor David Levine and three anonymous referees for valuable comments. Ignacio Palacios-Huerta and Jesse Shapiro provided helpful conversations. Phil Gordon was instrumental in our efforts to recruit world-class poker players. Omar Al-Ubaydli, David Caballero, Dwyer Gunn,

We are deeply grateful to the Jerusalem Media and Communication Center and the Development Studies Programme at Bir Zeit University for kindly providing us with their micro data. The authors thank Jesse Shapiro, Ivàn Fernàndez-Val and seminar participants at the Samuel Neaman Institute,

The findings and conclusions expressed are solely those of the authors and do not represent the views of SSA, any agency of the Federal Government, or the RRC. We thank Jesse Shapiro for early conversations, which encouraged us to write this paper, as well as Eric French and Randy Wright