While marriage remains the foundation of family life in the United States, the traditional process of family formation, specifically marriage before having children, has been dwindling. The proportion of American children born to unmarried parents has increased dramatically over the past three decades, from 12% in 1970 to nearly one-third of all births today (SigleRushton and McLanahan 2002). The decoupling of marriage and fertility behavior is particularly common among the low-income, less-educated urban population (McLanahan and Sandefur 1994). Unmarried parents often have fewer resources, and their children tend to display inferior outcomes compared to those raised by two married parents. (1)
Concerned with the rise in out-of-wedlock parenthood and its implications for the children involved, policies in recent years have been geared toward promoting marriage among unmarried parents. (2) However, little is known about the potential benefits of marriage after childbirth. Couples who have children out of wedlock are known to be selectively different from those who marry before having children. Unmarried parents tend to be of lower socioeconomic standing, face poorer prospects in the marriage market, and have lower incentives for assortative mating (Brown 2004; Garfinkel, Glei, and McLanahan, 2002; Rosenzweig, 1999). Hence, interpreting the outcome differences found in simple comparisons of children born to married versus unmarried parents as benefits of marriage can be misleading, as these differences may largely reflect the more favorable characteristics of married parents.
This study examines whether marriage between the biological parents after childbearing benefits the children involved, using data on a representative sample of children all born to unmarried couples drawn from the Fragile Families and Child Wellbeing Study (FFCWS). The FFCWS is particularly well suited to address this question as it provides child assessment data and detailed marriage, fertility, and socioeconomic information on both biological parents for a large representative sample of children born outside of marriage. We focus on the effect of marriage among parents who are romantically involved at childbirth on child cognitive ability measured at age 3, based on scores from the Peabody Picture Vocabulary Test (PPVT), a widely used interviewer-administered measure of receptive hearing and verbal ability. In addition, effects of marriage on child health and behavioral outcomes are analyzed.
A significant fraction of children in our sample experience the marriage of their parents. We analyze whether marriage after childbearing affects early developmental outcomes using an empirical strategy centering around a potential outcome framework similar to an experiment where the treatment (marriage after childbirth) is randomly assigned. We draw on matching methods (e.g., Heckman, Ichimura, and Todd, 1998; Rosenbaum and Rubin, 1983) to identify the treatment effect, exploiting the detailed information on the parents provided in the FFCWS. This approach addresses the selection into marriage by constructing an appropriate comparison group for children whose parents marry after childbirth. We first estimate the probability of marriage among unwed parents with a newborn, and then compare the outcomes of children whose parents share similar probabilities of marriage but differ only in whether their parents transitioned into marriage within 3 yr after their birth.
The present study also sheds light on the role of typically unobserved factors likely to be important determinants of selection into marriage, including the father's attributes and the degree to which the parents have similar characteristics (positive assortative matching). While some studies have examined the determinants of (marital) union formation among the population of single mothers (e.g., Aassve 2003), relatively little is known about the factors influencing the transition into marriage and the role of similarities in traits between unmarried biological parents. …