Thus, what appears to be a gradual decrease in fertility through time according to national-level statistics may actually be a process in which an increasing proportion of the population exhibits very low fertility.
The effect of education on fertility varied with the total fertility rate during the 20th century. For men and women born early in the century, who reproduced during the Depression and war years, education delayed fertility. There was little effect of education on fertility during the Baby Boom years, when all education groups showed a reduction in age of first reproduction. As education became an increasingly important determinant of wages in the last third of the 20th century and fertility decreased again, the effects of education on fertility were most pronounced (Kaplan et al., 2002).
Virtually all of the effects of education on fertility are due to postponement of marriage and a delay from marriage to reproduction. The delay, coupled with age effects on fecundity due to reproductive physiology, results in lower completed fertility. For example, data from the National Survey of Families and Households show that among women 35 to 44 years of age in 1990, high school dropouts have a mean of 2.77 children, which drops to 2.22 for those with high school degrees, 1.95 for those with bachelor’s degrees, and 1.43 for those with graduate degrees (Kaplan et al., 2002).
These shifts in the economy also appear to have had profound effects on mating and marriage. Two opposing forces act on marital stability. The decrease in complementarity associated with women’s entry into the labor force (especially when women began to reenter the labor force after short maternity leaves) renders divorce less costly, since women no longer rely solely on males for financial support. This change in the costs of divorce inspired the reform of divorce laws and, ultimately, increases in the divorce rate. On the other hand, the greater importance of education in wage determination increases educational homogamy in the mating market and parental assessments about the level of investment children require. Support for college is another form of balloon payment in the parental investment stream. Divorce has a large negative impact on both the probability that offspring will attend college and the likelihood of paternal financial support for college (Anderson et al., 1999b; Kaplan et al., 2002).
These two opposing forces affect people differentially, on the basis of their education and presumably their expected rate of return on investments in their children’s education. The effects of education on divorce when children are young are striking. Figure 7-5 shows the probability that a man will cease to live with a child before the age of 6 as a function of the child’s birth cohort and the father’s education level among men in Albuquerque, New Mexico. The increase in divorce rates is clearly evident. However, there is an interaction effect between a man’s education and the child’s birth cohort on the probability of separation between a child and its father. Among children whose father has less than a high school education,