(with Garance Genicot and Laura Mayoral). December 2025.
Summary. This paper revisits the Great Gatsby curve that connects inequality to mobility, using panel data spanning several countries and time periods. Existing literature observes that the intergenerational elasticity of earnings is positively correlated with inequality, implying that mobility (viewed as the negative of that elasticity) decreases with inequality. In sharp contrast, we show that measures of upward mobility, axiomatically based on progressivity in income growth rates, are robustly and positively associated with baseline inequality. While there is no logical contradiction here, our study highlights crucial differences between mobility measures based on (lower) persistence and those based on growth progressivity, and asks the reader to align their choice of measure with foundational criteria that they feel best describe “mobility.” Our findings offer a re-interpretation of the Gatsby curve through the lens of shared prosperity, and have implications for the evolution of inequality within countries.
We formalize the notion of collective dynamic consistency for noncooperative repeated games. Intuitively, we require that an equilibrium not prescribe any course of action in any subgame that players would jointly wish to renegotiate, given the restriction that any alternative must itself be invulnerable to subsequent deviations and renegotiation. While the appropriate definition of collective dynamic consistency is clear for finitely repeated games, serious conceptual difficulties arise when games are infinitely repeated.
International Journal of Economic Theory6 11–28, 2010.
Summary. The Phelps–Koopmans theorem states that if every limit point of a path of capital stocks exceeds the “golden rule,” then that path is inefficient: there is another feasible path from the same initial stock that provides at least as much consumption at every date and strictly more consumption at some date. I show that in a model with nonconvex technologies and preferences, the theorem is false in a strong sense. Not only can there be efficient paths with capital stocks forever above and bounded away from a unique golden rule, such paths can also be optimal under the infinite discounted sum of a one-period utility function.
