To explore the socioeconomic factors underlying decreasing marriages, we first look at changes in educational attainment of young Japanese women as compared to their male counterparts. According to existing studies (Raymo 1998, 2003; Tsuya 2009, 2012; Tsuya and Mason 1995), young women’s increasing educational attainment is a major factor that has caused delayed and fewer marriages in Japan since the mid-1970s. Table 5.1 shows changes, by sex, in the percentage of graduates advancing from junior high school to high school and on to junior college or 4-year college or university. Because Japan’s formal education system is highly competitive, with strong age barriers, people rarely have the luxury of alternating their commitments between education and other activities, such as employment. Economic and social disadvantages accrue for those who do not succeed in a series of structured “contests” at specific time points in their educational careers or at their entrance into the labor market (Brinton 1988; Tsuya and Choe 2004; Turner 1960). As a result, in Japan advancement to higher education and entry into the labor market tend to occur mostly within a narrow time frame, from the late teens to the early twenties.

We can see from Table 5.1 that the rate of educational advancement beyond high school increased sharply for women between 1970 and 1975, which is just before the onset of fertility decline to below-replacement levels. In the same period, high-school education became virtually universal.⁵ While a similar leap in the advancement rate from high school to university is seen for men in the same period, the male advancement rate shows moderate declines and stagnation from 1975 to the mid-1990s. Furthermore, whereas the female advancement rate to junior college peaked at around 25 % in the mid-1990s, the rate of females advancing on to university has continued to increase rapidly and steadily, reaching 45 % in 2010.

Given the rising advancement rates to higher education, the proportion of young women who are enrolled in or have completed junior college or 4-year college or university has increased sharply, with especially dramatic gains in the 1970s and 1980s. The proportion of women age 20–24 with higher education rose from 6 % in 1960, to 17 % in 1970, 40 % in 1980, and 60 % in 2010 (Sorifu Tokeikyoku 1964, 1973, 1984; National Institute of Population and Social Security Research 2014, p. 175). The proportion of women age 25–29 with higher education increased from 4 % in 1960 to 26 % in 1980, and 52 % in 2010.

These findings suggest that educational opportunities for women have expanded dramatically since 1970. Although the proportion of young men with higher education also increased after 1990 (and is still somewhat higher than that of young women), the increases among women have been steeper and more consistent.

How are these gains in educational attainment among young Japanese women related to their employment patterns? Table 5.2 shows the percentages of female and male graduates obtaining regular employment (full-time employment with job security and fringe benefits) within 1 year of graduation by educational level. The changing percentages from 1950 to 2010 reveal strikingly different trends for the two lower and two higher levels of education. For junior-high-school and high-school graduates, the rate of obtaining regular employment upon graduation declined precipitously in the 1970s and reached low levels in the 1980s and 1990s, presumably due to the rapid spread of higher education. In contrast, the rate of obtaining regular employment among female junior college and university graduates increased steadily from 1970 to 1990, with the gender gap among university graduates disappearing completely by 1990.

During the 1990s, the rate of obtaining regular employment upon graduation decreased dramatically for both sexes and at all educational levels. This was presumably triggered by the burst of Japan’s “bubble economy” at the end of the 1980s, which set the stage for fundamental changes in the nature of the Japanese economy and the types of jobs available to young people. In the early 1990s, in an attempt to increase their competitiveness and profitability in the face of economic globalization, employers began to move away from offering lifetime employment, a long-prevalent feature of the postwar Japanese labor market (Cargill et al. 1997, pp. 91–116). This resulted in rapid increases of temporary employment such as keiyaku (employment under a fixed-term contract with limited provision of social insurance) or haken (contract work whereby an agency sends workers to an organization for a specified period) (Igarashi 2009).

Workers with these temporary jobs tend to be young: 61 % of those employed as keiyaku or haken in 2005 were age 15–34 (Somusho Tokeikyoku 2005). Another notable phenomenon in the Japanese labor market is an increase of so-called “freeta,” young people who do not hold a stable job but rather hop from one temporary job to another (Somusho Tokeikyoku 2005). This deterioration of employment prospects for young people has also occurred in newly industrializing economies in Asia and in many industrialized countries in the West (Imai 1999; ILO 2011; Lee 2010; OECD 2002, pp. 129–69). As in other industrialized societies (Oppenheimer 1994; Oppenheimer et al. 1997), the proliferation of temporary employment has likely had a negative impact on marriage and the family in Japan (Igarashi 2009; Kohara 2010), particularly affecting the marriage prospects of young men with temporary jobs (Tsuya 2012).

To assess how young women and men fare once they enter the labor force, not just immediately after graduating from school, we next look at changes in age-specific labor-force participation rates by sex from 1960 to 2010. As shown in the upper panel of Table 5.3, the labor-force participation rate for women age 25–34 increased rapidly after the mid-1970s. Labor-force participation for women in their 30s still dips somewhat, compared with younger and older women, confirming the well-known M-shaped age pattern of Japanese women’s employment (Brinton 1988), but the drop in labor-force participation among women at peak childrearing ages has become much less notable in recent years. Although there is still a tendency for Japanese women to exit the labor force upon the birth of their first child and re-enter it once their last child enters school, it appears that this pattern now has less influence on labor-force participation rates, largely because of declining marriage rates among this age group. The labor-force participation rates for women in their late 30s and in their 40s has also risen substantially, implying that women at these higher reproductive ages are less likely to be married than in the past and, even if they are married, are more likely to be employed.