The Dynamics of Government: a positive analysis∗

How does the size of the transfer system evolve in the short and in the long run? We model income redistribution as determined by voting among individuals of different types and income realizations. Taxation is distortionary because it discourages effort to accumulate human capital. Voters are fully rational, realizing that transfers have implications also for future economic decisions and taxation outcomes. In our economy, our politically driven redistribution provides insurance, and we investigate to what extent the democratic process provides it appropriately. A general finding is that redistribution tends to be too persistent relative to what would have been chosen by a utilitarian planner under commitment. The difference is larger, the lower is the political influence of young agents, the lower is the altruistic concern for future generations, and the lower is risk-aversion. Furthermore, there tends to be too much redistribution in the political equilibrium. Finally, we find that the political mechanism is important: settings with smooth preference aggregation–we analyze probabilistic voting here–produce less persistence and do not admit multiple expectational equilibria, which occur under majority-voting aggregation. ∗We thank Costas Azariadis, Alex Cuikerman, and Joseph Zeira for comments. †John Hassler: IIES and CEPR, IZA and CESIfo; e-mail: [email protected]. ‡Per Krusell: University of Rochester, IIES, and CEPR; e-mail: [email protected]. §Kjetil Storesletten: IIES and CEPR; e-mail: [email protected]. ¶Fabrizio Zilibotti: University College London and CEPR; e-mail: [email protected].