The Japanese Context and Existing Empirical Evidence
To evaluate claims that empirical patterns based largely on data from a single country are broadly generalizable, two types of comparative research are useful. One involves gathering data from a large number of countries and comparing patterns across those countries. This can involve descriptive analysis (McLanahan 2004; McLanahan and Jacobsen 2015; Perelli-Harris et al. 2010; Raymo et al. 2015; Rendall 2010) or multilevel regression analysis (Harkonen and Dronkers 2006; Kalmijn 2013). This approach is powerful for its ability to describe or estimate relationships between the outcome of interest and contextual variables of theoretical relevance. The second type involves focusing on a single country (or small number of countries) characterized by distinctive features of theoretical interest, such as specific policies and distinctive social or cultural factors (e.g., Rendall et al. 2009). This approach is powerful for its ability to provide detailed insights into the specific conditions under which the general pattern of interest does or does not hold. In both cases, the key source of information is theoretically relevant contextual similarities or differences across countries.
Japan is rarely included in large cross-national studies of educational differences in family behavior (see Carlson et al. 2014; Raymo et al. 2015 for exceptions), but several studies have examined socioeconomic differentials in family behavior within the country (e.g., Fukuda 2013; Ono 2003; Raymo 2003). In contrast to the U.S. (and to a lesser degree, European countries), research on Japan that explicitly addresses the notion of diverging destinies is very limited. Indeed, a recent search of Google Scholar found only two Japanese language research papers that reference McLanahan’s (2004) paper. This limited attention to diverging destinies may reflect a view among Japanese social scientists that family change in Japan does not really fit into the larger second demographic transition framework. A few studies have addressed the question of whether Japan is experiencing a second demographic transition (e.g., Atoh et al. 2004; Matsuo 2001), but very low levels of nonmarital
© The Author(s) 2017
J.M. Raymo and M. Iwasawa, Diverging Destinies,
Population Studies of Japan, DOI 10.1007/978-981-10-0185-7_2
childbearing, the relatively low prevalence of cohabiting unions (Atoh 2001a), and limited evidence of “individuation” in attitudinal data (Atoh 2001b) may limit the perceived relevance of the second demographic transition framework and related questions regarding diverging destinies. As noted in the previous chapter, we view McLanahan’s (2004) explicit linkage of diverging destinies with the second demographic transition as suggestive, but not necessarily implying that family bifurcation will occur in tandem with the progression offamily changes typically associated with the second demographic transition. Our goal in this chapter is to provide an overview of both empirical and theoretical reasons to expect that family change in Japan may, or may not, be consistent with the patterns in research on diverging destinies or the “pattern of disadvantage” described in the previous chapter.
To accomplish this goal, we draw heavily upon our own work. Over the past decade, we have published several studies that articulate key reasons to believe that Japan is an unlikely setting for changes that conform to a pattern of diverging destinies. These include the long history of homogeneity in the family life course, limited use of modern contraception, and perhaps most importantly, a relatively strong maintenance of highly asymmetric work and family roles for husbands and wives. At the same time, our past research has recognized several features of the Japanese context that are potentially consistent with the kind of family bifurcation observed in the U.S. and many other low-fertility settings. Chief among these are relatively limited public spending on families and a rapidly changing employment environment that has had a very different impact on those at the upper and lower ends of the educational distribution, with implications for social and economic inequality.