Reexamining the Link of Early Childbearing to Marriage and to Subsequent Fertility

Citation: Morgan, S. Philip, Rindfuss, Ronald R. Reexamining the Link of Early Childbearing to Marriage and to Subsequent Fertility.
Internet Archive Scholar (search for fulltext): Reexamining the Link of Early Childbearing to Marriage and to Subsequent Fertility
Tagged: uw-madison (RSS), wisconsin (RSS), sociology (RSS), demography (RSS), prelim (RSS), qual (RSS), WisconsinDemographyPrelimAugust2009 (RSS)

Summary

Using data from the 1980, 1985, and 1990 Current Population Surveys (CPS), the authors investigate the relationship between early childbearing and unmarried childbearing and early childbearing and subsequent fertility. Earlier studies found a strong relationship between early childbearing and higher subsequent fertility. The CPS includes the date of the first four births and the last birth, data on the first two marriages and the most recent marriage, number of children ever born, and number of times married. Using this data, the authors construct a variable giving the legitimacy status of the first birth. Legitimacy status is coded into three categories: nonmarital (OW), legitimated, meaning that it occurred in the first eight months of marriage (L), and postmarital (PM). They include race as a stratifying variable (black or non-Hispanic white) and control for cohort of birth. The following variables are included in the hazard models of the pace of fertility: duration since first birth, period of first birth, age at first birth, legitimacy status, an interaction between duration and age at first birth, and race. The results show that the link between early fertility and nonmarital births has become stronger. The likelihood that a birth is nonmarital declines sharply as age at first birth increases. The odds are highest for blacks in the most-recent birth cohort (1985-1989) and lowest for whites in the earliest birth cohort (1950-1954). The odds that a birth is nonmarital have increased for all age and racial groups, but they have increased the most for those with an earlier age at first birth. At the same time, 'legitimation' of births has declined dramatically for all groups, but is greatest for the younger age-at-first-birth categories. The racial differential declines with increasing age at first birth and increases in the more recent periods. Women who give birth earlier are increasingly likely to be unmarried. In contrast, the authors find a weaker association between first births at young (versus older) ages and 1) a rapid pace of subsequent childbearing and 2) higher completed fertility. The 1965-1969 first-birth cohort is a 'watershed' group. Differences in mean parity between those bearing children at age 15-17 and those bearing children at ages 25-29 disappeared mainly during the period between 1962 and 1967. Declines in the pace of subsequent fertility are pervasive across duration, but are sharpest at the shortest durations. This indicates that fewer women are making the transition to second parity, but even among those that so make the transition, they make it at a slower pace. The results show a much slower pace of childbearing among those bearing children earlier, while older childbearers have experienced more modest declines. The timing difference between those with a marital conception and a premarital conception has disappeared, while the slower pace of fertility for those with a nonmarital birth has persisted throughout this period. For later cohorts or whites, the association between age at first birth and completed fertility is much weaker than in previous cohorts, the association weakened linearly over the period. For blacks, the association is weaker for early cohorts than for middle cohorts, but then attenuates again for the most recent cohorts.