In contrast to low-fertility societies in the West, the prevalence of nonmarital childbearing has remained negligible in Japan. The proportion of all births registered to unmarried mothers reached 2 % in 2005 after hovering around 1 % since the 1960s (National Institute for Population and Social Security Research 2015). Because the prevalence of nonmarital childbearing is so low, research on this pathway to family formation is limited. It is clear, however, that nonmarital births are much more common among women with lower levels of education. Tabulations presented by Raymo et al. (2015b) show that among first births occurring between 1995 and 2005, the proportion to unmarried mothers was 5 % for those who did not complete high school and vocational school graduates, 3 % among high school graduates, and only 1 % among four-year college graduates.
Despite the low levels of nonmarital childbearing, it is clear that relationships between pregnancy and marriage have changed markedly. This is most evident in the sharp rise in bridal pregnancy (marriages preceded by pregnancy). The proportion of first marriages that was preceded by pregnancy doubled from 10 % in 1980 to 19 % in 2010 (Iwasawa and Kamata 2014). This pathway to family formation is heavily concentrated among women with lower levels of education and this negative educational gradient has increased sharply over time. Comparing the 1970s and 1990s marriage cohorts, Raymo and Iwasawa (2008) showed that the probability of first marriage preceded by pregnancy doubled for women with a high school education or less while remaining stable at significantly lower levels for women who completed a two- or four-year college degree.
One way to think about this trend is to view the emergence of pregnancy as a primary reason for marriage in a setting where other incentives to marry (especially at younger ages) have waned. That is, couples postpone marriage until pregnancy necessitates formalization of the relationship. The link between marriage and fertility remains strong, but the temporal ordering of marriage and pregnancy has reversed for many couples. This scenario would be consistent with the notion of diverging destinies if bridal pregnancies are associated with lower quality, less stable marriages. However, it would be less consistent if bridal pregnancies simply hasten stable marriages that would have occurred sooner or later. Carefully distinguishing between these two scenarios is an important task for future research.
The limited empirical evidence that is available appears more consistent with the first scenario. Raymo and Iwasawa (2008) found that the patterns of educational pairing in marriages preceded by pregnancy differ significantly from those not preceded by pregnancy, with women who are pregnant at the time of marriage much more likely to marry a man with less education than themselves. It is also clear that bridal pregnancies tend to be reported as unintended. Raymo et al. (2015b) find that the higher prevalence of unintended childbearing (among first births) for women with a high school education or less is partially explained by the fact that these women are more likely to have a first birth that was the result of a premarital pregnancy. Because female educational hypogamy and unintended childbearing have been linked to subsequent marital instability and lower levels of well-being, growing educational differences in the relationship between pregnancy and marriage may play an important role in the intergenerational transmission of disadvantage. We examine trends in educational differences in nonmarital childbearing and bridal pregnancy in Sects. 4.3 and 4.4.