Group Effect, Productivity and Segregation Optimality

In literature on residential segregation (RS hereafter) there are two important results about the social optimality of this phenomenon. Firstly, the maximum level of RS can constitute a social optimum if one part of the population generates negative externalities on the remaining one. The population suffering the negative externalities can be called a prejudiced population. Under these circumstances traditional bid-rent models with externalities and general equilibrium models have showed the optimality of RS 3. On the other hand, when the individuals’ preferences are to live in balanced neighborhoods, high levels of RS diminish the aggregated utility, consequently, full integration is the social optimum. The well known Schelling’s model has stressed this issue, showing how a population, in an artificial world, can evolve to a segregated society although individuals want to live in a perfectly balanced neighborhood, reaching a bad, but the only stable, equilibrium. If the prejudiced population coexist with a population preferring balanced neighborhood, the literature has proposed the payment of compensating transfers by the prejudiced population to the non-prejudiced ones to accept the exclusion (Anas, 2002).