We present a model of the creation of social networks, such as political parties, trade unions, religious coalitions, or political action committees, through discussion and mutual persuasion among their members. The key idea is that people are influenced by those inside their network, but not by those outside. Once created, networks can be "rented out" to politicians who seek votes and support for their initiatives and ideas, which may have little to do with network members' core beliefs. In this framework, political competition does not lead to convergence of party platforms to the views of the median voter. Rather, parties separate their messages and try to isolate their members to prevent personal influence from those in the opposition.

We estimate how 1970–90 changes in an outcome‐based measure of school quality (state average test scores) affected changes in earnings for those leaving high school to enter a state’s labor force. We find that a one standard deviation deterioration in a state’s relative test score performance is associated with a 3% (or .5 SD) reduction in average wages of young entrants to the labor force. We also find a similar decline in college matriculation. There is weak evidence that the school quality effect on earnings diminishes as labor force entrants acquire experience.

We analyze entry, pricing, and product design in a model with differentiated products. Market equilibrium can be “separating,” with multiple sellers and a sorting of heterogeneous consumers across goods, or “exclusionary,” with one seller serving all customer types. Entry into an initially monopolized market can occur because of cost reductions or product improvements, but entry need not lower the incumbent’s price, improve efficiency, or raise consumer welfare. Postentry design incentives favor a softening of price competition and stronger market segmentation, whereas exclusionary design changes typically raise consumer welfare. Potential, as distinct from actual, entry always benefits consumers.