Saving rates have fallen steadily in the U.S. since the 1970s. The literature has proposed numerous possible explanations for the decrease including increases in governmental insurance, changes in the distribution of income, and increases in annuitized wealth. Analysis of all three explanations is complicated by dramatic increases in married women’s labor supply and rates of marital disruption during the same period. These changes have significant effects on the precautionary and consumption smoothing motives for both singles and couples. For singles, saving affects and are affected by the prospect of marriage. For couples, the prospect of future marital disruption affect saving directly and through its impact on the labor supply decisions of household members. Moreover, the amount of saving accumulated by the household modifies the probability of marital disruption by changing the outside options available to the household members. The purpose of this paper is to examine the interactions between marital status, labor supply and wealth accumulation using an intertemporal collective model with limited commitment. In the proposed framework, married couples choose efficiently the optimal amount of consumption, leisure and saving until one or both agents are better off being single for at least one period. In this event, if there is some surplus to be divided, the two spouses will renegotiate the intra-household allocation of resources and stay married, otherwise they will split and be single for at least one period. Single agents choose consumption, leisure, saving and whether to marry a potential partner drawn from the population of singles. The number of children in a household is modelled as an exogenous process. The framework is estimated by simulated maximum likelihood using the 1984-1993 waves of the PSID and the 1984 and 1989 wealth supplement files. The ten-year panel provides information on the timing of childbirth, marriage and divorce as well as yearly data measuring the flows of child-support, alimony and voluntary transfers between divorced individuals. The wealth supplement files contains detailed data on wealth accumulation, which are crucial for the estimation of the proposed model
