1 The changing Japanese family
Marcus Rebick and Ayumi Takenaka
The 1990s saw major changes in the way in which the Japanese think and talk about themselves. One area where this is manifest is in discourses about the family. The family is in âtransitionâ (Imamura 1990; Ochiai 1997); it is in âcrisisâ, (Hayashi 2002); and other times, it is simply described as becoming more diverse in its forms and definitions. In contrast to an idealized nuclear or extended family, there are many more families today (White 2002). There are so many types of families that they could even be described as âindividual networksâ (Fukushima 2002). How has the family changed, or remained unchanged, and why? To what extent is this unique to Japan? And what implications does it have for Japan as well as other industrial societies?
In the context of economic slowdown and major demographic shifts in the 1990s, the Japanese family has undergone a number of significant changes. The marriage rate has fallen, while the divorce rate has risen. The patterns of cohabitation of members of the extended family have also shifted with the decrease in âtraditionalâ three-generation households and with the increase in the number of young men and women who continue to stay single and live with their parents. Meanwhile, the mean age of first marriage has climbed to the worldâs highest level, and fertility has fallen to a level well below replacement. Japan is the most rapidly ageing society in the world, and the population is projected to decline after 2006. This rapid ageing of the population will necessarily imply great change in intergenerational relations, as there will be far fewer young to look after the old. It will also have significant ramifications for the nationâs social welfare and labour market. These issues are at the heart of public debates in Japanese society today.
Rapid social transformation accompanied by population ageing, plummeting fertility rates and diversified definitions of the family are, indeed, common among many industrial countries. Italy, in particular, faces similar problems and has many characteristics that invite comparison with Japan. Both countries have low divorce rates and low rates of birth outside marriage. In both countries, children tend to live with their families until marriage. Both countries have an industrial structure characterized by many small firms and family-run enterprises. Both countries experienced defeat in the Second World War but recovered with rapid economic growth in the 1950s and 1960s. This is part of the reason why both countries have had rapid demographic transitions to low fertility in the early post-war period, leading to an ageing population structure in the early twenty-first century. Both countries also began to receive growing numbers of immigrants (foreign migrants) in recent decades. (As of 2002, 2.0 million and 1.5 million foreign residents were registered in Japan and Italy, making up 1.6 per cent and 2.6 per cent of the nationsâ populations, respectively SOPEMI 2004).1 These numbers, however, remain fairly small in comparison to other industrial countries, although immigration is increasingly seen as a way to âsubstituteâ for the decreasing and ageing populations (United Nations, Population Division 2000).
Moreover, both countries appear to have similar approaches to the provision of social welfare. Esping-Andersen (1990) places Japan in a corporatist/conservative group along with Italy in his scheme of âthree worlds of welfare capitalismâ. In this view, Japan, along with Italy, is typical of countries that are âstrongly committed to the preservation of traditional familyhoodâŚ(t)he state will only interfere when the familyâs capacity to service its members is exhaustedâ (ibid.: 27). Using Italy for a comparison, we come to see that the changes occurring in Japan are representative of a pattern common to other countries, especially those of Southern Europe, rather than an aberration based on âuniqueâ features of Japanese society.
This volume examines the changes that are taking place in the Japanese family comparatively by looking at broader social, economic and global contexts. We begin by looking at the demographic changes including the rising age of marriage, declining birth rate and ageing of the population and their effect on social welfare arrangements in Japan. It is no longer considered desirable or even feasible to leave the care of children or the elderly entirely to the family. The demographic change is also compared to similar trends in the south European countries. The next part of the book looks at the gender balance of labour and responsibility both within the household and at work, including the way in which the Japanese language reflects these shifts in the gender balance. The third and final part of the book explores the changing definition and function of the family. There has been a shift in the boundaries between family and community and between family and state, and the role of the state in the provision of social welfare has increased. Here, we also find a useful comparison with developments in Italy.
Before beginning our discussion, we would like to clarify some of the terms that we will be using in this introduction. The term setai refers to a household, officially defined in Japan as a group of people sharing a kitchen. The term ie, usually rendered in English as âstem familyâ, refers to a family lineage with a (theoretically) eternal existence, stretching from the distant past into the future. The ie, which was institutionalized in the Meiji period in an attempt to modernize the nationâs family system, is preserved through patrilineal links between generations. Typically the eldest son inherits responsibility for the ie, taking over as its head. All of his siblings will either marry out of the ie (in the case of women) or move out (in the case of younger brothers). Reinforced by a patriarchal head and hierarchy by birth, the ie system maintained a strict gender-based role division within the family and household. What makes the Japanese ie distinctive, however, is the widespread use of adoption whenever the survival of the ie is at stake. If there are no sons, or if none of the sons is suitable to be an heir, the husband of a daughter may be adopted to become the new head of the ie. If there are no children at all, then a son may still be adopted. An important feature of this system, therefore, is the continuity of the ie, or family line, rather than an emphasis on blood relations. Japanese law, since the beginning of the Meiji period, has underpinned this system by stipulating that all Japanese belong to the ie (thus, non-Japanese residents do not belong to one), as reflected in the population registration system. Known as koseki, this registration system ensures that all married couples and their children share a surname as a family unit. The vast majority of the time, women change their names in marrying or moving into their husbandsâ registries. Finally, the term kazoku is typically translated as family and broadly understood as blood-related kin and conjugal relationships, while katei, or âhomeâ, broadly refers to the âfamily domainâ or a place where kazoku shares a communal life. Kazoku is the most amorphous and the hardest to define, and some use the term to refer to peopleâs sense of family, or family identity (Ueno 1994). Murdock (1949) defined the family as âa social group characterized by common residence, economic cooperation, and reproductionâ. The family, since then, has diverged so much in its functions, types, and meanings, and it is difficult to define it simply by these characteristics (Tsumura 2002). It may be more useful to think of âthe familyâ in terms of boundaries.
Boundaries are shifting for all these â the concepts of setai, ie, kazoku and katei. Before we discuss these shifts, we must look more closely at the demographic and socio-economic changes that are taking place, and how they are related to the state of the family.
Japanâs demographic transition
Ageing
Japan, followed by Italy, is the most rapidly ageing society in the world. Figure 1.1 shows trends in the proportion of the population over the age of sixty-five over time in comparison with some other OECD countries to illustrate this point. In 1950, the proportion of the population aged 65 and over represented only 5 per cent of Japanâs population; by 1997, this rate increased to 15 per cent, surpassing the proportion of children under 15, and increased further to 20 per cent by 2005. The proportion of the elderly over 65 years of age is projected to increase further to 22.5 per cent in 2010 and 30 per cent in 2030 (CAO Cabinet Office 2004a). The causes of this phenomenon are well known and reviewed by Ogawa et al. in this volume. Japan moved through what is usually called the âfirst demographic transitionâ rapidly in the late 1940s and early 1950s, with the total fertility rate (the number of children born on average to each woman over her lifetime) dropping from 4.54 in 1947 to 2.04 in 1957 (MHLW1 2004). At the same time, neo-natal mortality rates dropped due to better health care and nutrition. There is nothing exceptional about this pattern, common to all economically developed societies, but its rapidity has resulted in the rapid ageing of the population that we see today. The other factor that has been important is the great increase in longevity of the Japanese population. In 2003, Japan had the highest life expectancy in the world, at 78.4 for men and 85 for women (Statistics Bureau 2004). Life expectancy has increased consistently over the period since the Second World War, rising by nearly 30 years since 1947, but half of this rise occurred in the 13 years up to 1960. In the years since 1960, life expectancy rose more gradually, but this is the period in which Japan overtook other developed countries.
Figure 1.1 Percentage of population over age 65, selected OECD countries.
Source: United Nations, Population Database.
Cohabitation patterns
Along with ageing, Ogawa et al. (Chapter 2, this volume) point out that the patterns of cohabitation have shifted with the decrease in three-generation households, both in terms of absolute numbers and share of all households. In 1975, three-generational households represented 18 per cent of all households, but the percentage decreased gradually to 10 per cent in 2002, while the share of single-member households increased, from 18 per cent in 1975 to 23 per cent in 2002 (MHLW1 2004). Elderly households, made up of members over 65 years of age, also grew in numbers, constituting 15.8 per cent of all households in 2003. Overall, the average number of household members steadily decreased, from 5 in 1950 to 2.5 in 2004 (ibid.).
The meaning of cohabitation, however, brings us to another of the ways in which the boundaries of the family have become blurred. In recent years, many families have resolved the difficult problem of care for the elderly by developing arrangements where the older generation lives next to the younger ânuclearâ household, but not directly with it. One example of this may be seen in cases where property is subdivided and the younger family lives in a separate house next door. Another arrangement is to have the house itself subdivided with a separate apartment for the older generation to live in (Brown 2003). Bettio (Chapter 4, this volume) reports a similar trend in Italy where many grown-ups continue to live near their parents after moving out of their parental homes. At the same time, living nearby does not automatically translate into close family relationships; Japanese children, who live with, or near, their parents more frequently than American and Swedes, for instance, reportedly maintain less frequent contact with their parents (Inoue and Ebara 2005).
While these arrangements may not technically be defined as cohabitation, it would be difficult to classify them as completely separate households. A question that arises is how to define a household (setai), or how to classify a variety of arrangements, including the cases where different generations live in close proximity in the same city, rather than separated by great distances. From a policy viewpoint, it is important that statistics reflect this variety as much as possible, taking into account the blurring of what in the past may have been seen as a clear distinction.
Falling fertility
Another significant demographic shift over the last ten years has been the decline of the fertility rate to levels well below replacement level. In 2004, for example, it is estimated that the total fertility rate stood at 1.29 whereas the level needed for maintenance of the population is 2.1 (MHLW2 2005). Although the decline in fertility first became a major issue in the 1990s, the actual decline from replacement levels began in 1973 and has continued steadily ever since. This decline in fertility is usually distinguished from the earlier decline as there was a period of stable fertility from the mid-1950s to the early 1970s. The earlier decline was caused by the new availability of contraceptives (mainly condoms) and the legalization and provision of abortion. At the same time, falling levels of infant and child mortality would have meant that families needed fewer children to be sure of reaching a desired family size.
The more recent decline of the last thirty years, sometimes referred to as the âsecond demographic transitionâ, is largely attributed to improved employment opportunities (including higher wages) for women, along with the rapid growth in the educational attainment of women. In 1970, the average woman aged 25â29 working full-time as a regular worker earned only 62 per cent of what her male counterpart earned.2 By 1999, she was earning 83 per cent.3 The percentage of managerial posts held by women tripled between 1982 and 2003 (Rebick, Chapter 5, this volume). In 1970, 18 per cent of high-school graduate women were going on to higher education, and most were headed to two-year colleges. In 2003, 47 per cent were headed to higher education (compared to 42.7 per cent of men), and almost two-thirds of these went to four-year universities (MOE 2004).4
The declining fertility rate is also due to perceived difficulty in raising children, particularly by working mothers. As Nagase (Chapter 3, this volume) points out, many women perceive that it is difficult to balance work and childrearing, and this is, in part, attributable to a perception of inadequate childcare facilities and high costs associated with childcare (CAO Cabinet Office 2001a). Consequently, the ideal number of children (2.62 in 1982 and 2.56 in 2002) far exceeds the number of children women actually have (Inoue and Ebara 2005).
Declining marriage rate, delayed marriages
Rising aspirations and the rising opportunity cost of marriage have led to an increasing average age of (first) marriage â 29.6 for men and 27.8 for women (MHLW2 2005) â which is associated with a drop in the number of children born within marriages. (Out of wedlock births remain very low in Japan, at 1.9 per cent of all births in 2002, compared to 11 per cent in Italy, 56 per cent in Sweden, 41 per cent in the United Kingdom and 34 per cent in the United States. (Bettio, Chapter 4, this volume; Martin et al. 2003). They have also led to a dramatic rise in the proportion of women who never marry (Ogawa et al. Chapter 2, this volume). On average, the drop in the number of children within marriage and the rise in the proportion of women who never marry have contributed about equally to the decline in fertility (Retherford et al. 1996). Much of the decline in the numbers born within marriage can be attribu...
