We will study both the determinants of marital dissolution and its short- and long-run consequences for children. We propose an integrated model that includes the determinants of marital dissolution, the division of resources between parents and children and among children in intact and non-intact families, and the impact of dissolution on short- and long-run outcomes for the children. We consider both short-run child outcomes such as test scores shortly after dissolution, and longer-run outcomes such as duration of schooling, earnings and occupational status which usually have not been considered before. We extend previous intrahousehold allocation theoretical models in two directions: we explore the allocation of resources between parents and children and we model parental concern about the distribution of inputs among children as well as outcomes. We use latent variable techniques and sibling data to control for unobserved child-specific (e.g., genetic endownments) and family-specific factors (e.g., discord) in the entire system of relations. This enables us to explore the extent to which the impacts on children are from actual marital dissolution rather than factors that lead to high propensities for dissolution. Furthermore, such techniques will enable us to develop better measures of unanticipated events (""""""""surprises""""""""). We will use four rich data sets that will permit considerable advances of our understanding: the NAS-NRC Twin and Adult Offspring Sample, the Michigan Panel of Income Dynamics, The Fels Longitudinal Sample, and the Wisconsin High School Senior Sample.
