Testing mobility models: A rejoinder to Erikson and Goldthorpe

Despite high rates of cohabitation before marriage, and considerable efforts invested in the search for a life partner, a high number of marriages end in divorce. This paper develops a model of household formation and dissolution in which it may be rational for individuals to marry, fully anticipating that they will subsequently divorce. Economies of scale associated with living as a couple rather than in two separate households provide an incentive to marry; problems with free riding in the provision of household collective goods may lead to divorce. Marriages which involve partners who are similar, in tastes or in their productive capacities, and in which private goods are equally shared, are the most likely to be stable. In contrast, marriages which involve very disparate partners, or which share the fruits of market labor very unequally between the partners, are more likely to be shortlived and end in divorce.