Gender roles and social policy in an ageing society: the case of Japan

This article reviews the major underpinnings of the Japanese welfare state in the context of social care from a feminist perspective. In Japan, familycare responsibilities have traditionally been assigned to women; hence, care has long been a women’s issue. However, as the social contract of a male breadwinner and a ‘‘professional housewife’’ gradually fades out, Japanese women find more opportunities to renegotiate their caring roles. Of course, this social transformation did not occur in isolation, it was influenced by patterns in economic development, state policies and mainly demographic changes. All this has stimulated new state responses in the form of social welfare expansion that arguably aim to relieve women of the burdens of family-care. The issue remains, however, as to whether Japan would be able to recognise that the main structural issues of population ageing do not originate from demographic changes, but from a strict gendered division of labour and gender inequality.


Introduction
As the twenty-first century advances, Japan finds itself at a crossroads; only a few decades ago it was amongst the youngest societies in the developed world, now it is the oldest one (Long et al. 2009;Peterson 2002). In this regard, the analysis of the Japanese case could provide a relevant frame of reference for other countries as they also implement new social policies in response to population ageing and changing gender relations. Certainly, Japan's demographic situation originates in the convergence of two factors: increasing life expectancy and declining fertility rates. However, these demographic patterns are not the issue on their own. The real issue lies instead in the correlation of such patterns and the dynamics on gender relations and societal/individual views on care responsibilities, which altogether influence social policy responses. Throughout history, the Japanese state has been mainly focused on its industrial and economic growth leaving, therefore, little room for the development of social policies. Consequently, the Japanese welfare system has relied on the informal care families provide to their members, especially the elderly. Traditionally, Japanese women have been encouraged and expected to take on care responsibilities, whilst men fulfil their role as breadwinners. Today, however, the increase of women's participation in the workforce and the decrease in Confucian-inspired family values have together created significant changes in the gendered division of labour (Long & Harris 2000) as some women have started to question, or even refused to take on, the traditional caring role. In this sense, societal and individual changes in post-industrial Japan have made possible not only the establishment of new social care policies but highlighted the need for a social contract fairer to women.
For the theoretical framework, this article draws upon Daly and Lewis' (2000) notion of social care as a useful tool to analyse contemporary welfare states; particularly focused on explaining how care and gender issues are interlocked with welfare state development and social policymaking. Most of the feminist approaches on care have concentrated on the everyday aspects of care and how those condition women's wellbeing (i.e. micro level). However, a recent approach casts gender and International Journal of Ageing and Later Life social care responsibilities in a prominent place for welfare state policy (Daly & Lewis 2000;Gottfried & O'Reilly 2002). In their model, Daly and Lewis (2000) identify three dimensions for social care: (1) care as ''labour'', (2) care as an ''obligation and responsibility'', and (3) care as ''financial and/or emotional costs and benefits''. They note that by having a care-centred approach in welfare state analysis, one can uncover the intersection of private/public and formal/informal domains. Thus, the division of care as labour, obligation and costs within and between the family, the market and the state is the departure point that allows an analysis of both the micro and macro levels. Accordingly, Daly and Lewis (2000: 289) identify three types of European welfare state responses to the issue of care. For instance, they consider the Scandinavian states as a distinctive ''cluster'' as they tend to collectivise caring for both the elderly and children. The countries from the middle and south of continental Europe are identified as favouring ''privatisation'' of care, although with some degree of differentiation (given by their either Bismarkian or Mediterranean welfare models). The third type, identified as the Beveridgean welfare states, makes a strong distinction between child and elderly care, and also favours child care as a private rather than a public responsibility. More importantly, the authors suggest that the concept of social care could contribute to a much broader analysis in three specific lines: (1) by looking at how social policies are gendered; (2) by examining the linkages between specific policy domains (e.g. interrelation of elder and child care); and (3) by uncovering trajectories of change in welfare states. Despite this useful conceptualisation of social care in relation to welfare states, something is missing from this model. According to Peng (2002), Daly and Lewis' model should also account for the causes of specific welfare state responses, along with political economy dynamics between the family (i.e. women), the market and the state concerning the issue of social care. Figure 1 represents my interpretation of Daly and Lewis' model of care, along with Peng's contribution.
By following a feminist approach, this article aims at highlighting three main issues: (1) how societal changes, in the context of rapid ageing, have increased opportunities for women to negotiate their traditional  (2) the role of Japanese women's social and political discourse Á particularly during the 1980s and 1990s Á in shaping welfare state policies and their contribution to redefine care as a social issue; and (3) whether current Japanese policies are a real shift towards an expansion of public provision to diminish the burden care on women and balance work and family life, and ultimately promote gender equality. This article departs from the description of the ''unique'' characteristics of the Japanese welfare state. The second section will explore how recent demographic, social and economic factors have contributed to changing gender relations and eventually drawn attention to the relevance of social care in welfare state development. The third section will explore two specific social policies (i.e. the long-term care insurance and the promotion of a gender-equal society) with which the government has tried to respond to the challenge of an ageing society. Finally, the last section highlights the relevance of the Japanese case for anticipating the future of welfare states in ageing societies.

Uniqueness of the Japanese Welfare State
Given the processes of economic globalisation and the emergence of postindustrial societies, welfare states have adjusted to a changing environment and, therefore, cannot be categorised within a static and theoretical typology. Although most welfare state analyses have placed Japan into a Western framework, they do so by stressing its ''uniqueness'' for not fitting comfortably into any of Esping-Andersen's (1996) regime types (i.e. liberal, conservative and social-democratic). Yet, Japan could be placed closer to the conservativeÁcorporatist regime, along with Germany (Gottfried & O'Reilly 2002) and other Continental European countries, in which social rights are based on employment and contributions, and welfare responsibility is placed on families Á or community Á rather than the state. However, neither welfare state regime theory nor the East-Asian model or ''Confucian welfare state'' allows the Japanese welfare state to be easily situated (Izuhara 2003b: 1).
Thus, one could argue that Japan is a combination of the liberalÁ residual and the conservativeÁ corporatist regimes, but at a low level of development. In fact, it is this ''residual'' status of Japan's system, which helps classify it as a ''low-performing welfare state'' (Hill 1996) or even as ''an embryonic form'' (Bryson 1992) characterised by the weakness of its institutional form that relies on the private sector (family, community and corporations) and above all, gives priority to economic and industrial development, hence allocating low levels of public expenditure for social policy and welfare. (Campbell 2002;Eto 2001;Peng 2004). In fact, it is this emphasis on the economy over social redistribution that serves as the framework that identifies Japan as a productivist welfare state in an attempt to incorporate it into the Western logic of welfare development (Holliday 2000). Nonetheless, as shown in Table 1, each one of Esping-Andersen's regime types can be identified in a different period of Japanese social policy development, defined by key problems and the corresponding government actions. What is of greater interest is to uncover how traditional elements of Japanese culture, such as ie ideology, corporate-centred system, family-centred care and the nurturing/subordinated role of women, interacted in order to shape a particular type of welfare regime. Since Daly and Lewis' model denotes that a welfare state cannot be fully grasped unless gender and social care are integrated into the analysis, the focus here will be in those variables.

Characteristics of Japan's Welfare State
This section elaborates on what I consider to be the basic features of Japan's welfare state in relation to care. I contend that the state has strengthened and promoted traditional elements as unique to Japan by legitimising them as inherent to Japanese culture.
Certainly, beneath Japan's early welfare development lies the continuation of the ie ideology; a patriarchal system embedded in the role of family, gender relations and a hard-working corporate culture, that even today, as I will further elaborate, touches the life of every Japanese in a fundamental way. The ie family system was prescribed by the Civil Code during the Meiji Era, became to be known as the ''modern Confucian family'', and was indeed influential until the end of Second World War. A relevant aspect is that under this system, the family Á typically a threegenerational, patrilineal and patrilocal household Á was expected to provide support and care for its elderly members. In this family system the International Journal of Ageing and Later Life  (Campbell, 2002: 7).
Gender roles and social policy in an ageing society eldest son, or alternate heir, 1 exclusively inherited the family name, assets, social status, worship of ancestors (i.e. tomb, home Buddhist altar), and occupation (Izuhara 2003a: 85;Kono 2000: 182;Koyano 2000: 201Á202) and he, ideally, in return would provide material and affective care to his elderly parents not only out of obligation but as an actualisation of filial piety. 2 Although under this system the father was the head of the household and had full responsibility for the care of all members, it was the wife or daughter-in-law who provided all practical care, thus allowing the social construction of caregiving as a woman's primary work and as an uncontested extension of her domestic responsibilities. Besides originating the cultural meaning of caregiving in Japan, with intergenerational co-residence as a basic feature, the ie system also contributed to the country's rapid industrialisation. As only oldest sons could become their father's heir the rest of the family sons were therefore obliged to join the labour force or to found enterprises. This, however, is a thesis that has not always received sufficient discussion (North 2009).
As shown in Table 1, throughout the development of Japan's social policymaking and law the state has defined the role of the family, and subsequently, since pre-and post-war periods, that of private companies as the main welfare providers. Although, in theory, the necessity for welfare provision by other sectors increases with the expansion of modernity, there is the possibility that traditional values and culturally rooted customs are intentionally preserved and utilised politically. Certainly, since the post-war period the Japanese Government has not only relied mostly on the informal role of families (along with companies and communities) to provide social welfare, but has legally enforced such role as an obligation. This therefore explains why social welfare support rates in Japan have remained relatively low, when compared with other advanced countries. 3 Between the 1950s and the 1970s, credited as Japan's period of rapid economic growth, under a liberalÁresidual welfare regime the provision of social support to the family was considered a social cost for economic development. As Izuhara (2003b: 9) notes, during the 1970s the public increasingly put pressure on the Japanese welfare state. Finally in 1973, the programme ''Welfare Year One'' (Fukushi Gannen) aimed to make Japan a welfare state, striving for more welfare and a better life quality. However, the same year the Japanese economy was affected by the world oil crisis, followed by a serious stagnation period (Kono 2000: 183). Furthermore, 1970 was a very significant year for Japan: the country met United Nations criteria for an ''ageing society'' since people aged 65 and over comprised 7% of the total population (Tamai 2003: 41).
Subsequently, the government was intent on formulating new policies in an era of low-economic growth and a consensus materialised: the ageing of society should be a major national concern (Campbell 1992). Gradually, the idea of Fukushi Gannen gave way to slogans, such as ''reconsider welfare'' and ''welfare state disease'' (Izuhara 2003b: 9). Individual responsibilities were re-emphasised, by re-establishing traditional values of mutual aid within the family, communities and workplace. In order to control any increase in public expenditure, the family was explicitly understood as a partner and the core source of welfare provision (Harada 1988). With such re-definition of the family, by 1979 the government officially confirmed the residual principle of welfare to guide social policymaking accordingly, hence establishing the ''Japanese-style Welfare Society'' (Kono 2000: 183).
One explanation for the reappropriation of familism by the Japanese Government is, Miyamoto (2003: 13) would argue, that in Japan, in a situation similar to other advanced industrial societies, a large number of female workers withdrew from the labour market after Second World War and increased the number of full-time or ''professional housewives'' (Vogel 1978). The reason for this was the favouritism of the employment and social security systems towards male breadwinners. Consequently, retired female workers played the role of carers for children and elders. Hence, it is through various employment policies that familism in Japan came to occupy a major role in the social construction and reproduction of the traditional care dynamics and intergenerational responsibilities, thus cementing the dominant role of the family in elderly care provision, the gendered division of labour and subordination of Japanese women to family and men, while at the same time helped Japanese policymakers shaped a particular type of welfare regime framed by a limited social provision.
Finally, since priority was given to policies for economic and industrial development, the so-called ''uniqueness'' of Japan's economic miracle has been, Peng (2000) would argue, confused with a successful welfare system. The 1990s, however, witnessed the end of rapid economic growth and came to be known as ''the lost decade of Japan'', but it also marked the beginning of the rapid ageing of its population. Consequently, the socalled ''unique'' Japanese model underwent a more critical assessment and emerged as a recipe for disaster. In this regard, Campbell (2002) suggests that the practices associated with the Japanese-style of welfare, such as seniority and lifetime employment, are key factors causing economic stagnation. I would, however, add the ideology of familism as contributing to Japan's welfare crisis. This ideology is built on the notion ''[ . . .] that a full-time [female] caregiver will be [ever] present in the home to nurture the nuclear or extended family [ . . .]'' (Roberts 2005: 176). This is another factor that kept women out of the labour market. Consequently, one could argue that the Japanese state has faced similar issues as in the West, as it built its employment system and family structures under a patriarchal ideology with a gendered division of labour.
Indeed, through Confucian-embedded familism, the role of the Japanese state has been to allow and encourage family caring, which is similar to the normality of family care found in European welfare states still Á strongly Á influenced by Catholicism (e.g. Italy, Spain). Thus, is there definitely a ''uniqueness'' about Japan's welfare state? Certainly, Esping-Andersen's typology has proven to be a useful comparative framework that considers Japan as a mix of all three welfare regimes, which led scholars to appreciate Japan as a ''unique'' model. Importantly, in most European countries there seems to be a shift towards a mixed economy of welfare, where the state is transferring its role as welfare provider into the family and voluntary sector, whilst the market is also assuming a more active role in the provision of care (Daly & Lewis 2000). Although Japan is undergoing a similar ''mixed'' transformation, the difference is that the state's role has, historically, been more of a regulator than a provider of social welfare. Thus, Japan appears to be making a policymaking shift, which consists not only of In this sense, Japan's focus on social care policies appears to be informed by neoliberal expansion principles shifting also responsibility from ''the family'' to ''individuals'', and thus emphasising the rhetoric of ''selfresponsibility'', a concept that often evokes Western [i.e. American] ideas and practices of care tending towards a free-market model. Certainly, the problem to situate the Japanese welfare state as either unique or a mix of regimes derives from too many factors that distort its trajectory of development. It is on this basis that the welfare regime typology is insufficient to understand the dynamics of change in the Japanese welfare state. It is only when we apply a model that takes into consideration variables, such as issues of gender and the responsibility of care, like Daly Gender roles and social policy in an ageing society and Lewis' that we can begin to grasp the trajectories of development in Japan's welfare model and the significance of care as a domain in policymaking.
Therefore, when the focus is on care, it is easier to spot singularities amongst welfare states. For instance, whilst in most European countries there is a tendency for family policy to become care policy, in the case of Japan, arguably, there is no family policy as such (Roberts 2005;Schad-Seifert 2006), but instead it is elderly care what has maintained a significant place in the trajectory of development of the Japanese welfare state for the last 40 years (Campbell 2000). This may show then how the changes Á and responses Á taking place in Japan are one step ahead of most Western countries; obviously this emphasis on social policies for older people, is also due to the Japanese state's aim to balance demographics and economic development. Thus, instead of analysing cash transfers or taxed-services within welfare state transformation, the focus is on whether care has become a main concern for welfare states and social policy.

Pressures for Change
As indicated earlier, the Japanese welfare state has mostly relied on the family and corporate provision of care. In such context, Japanese women's family-care obligations continue to be reinforced by social and employment policies. However, such welfare regime, embedded in a strong male breadwinner/full-time housewife family model, is being challenged by demographic, social, political and economic trends. Moreover, such trends have created the opportunities for women to negotiate their traditional role, but have also acted as pressures changing the forms and conditions in the demand and provision of social care between the state, family, community and market. In this section, I present some of those trends looking first at the contribution of women's liberation movements to Japanese social policy before considering the current changes in women's care responsibilities at home and increasing opportunities outside the home.

Women's Socio-Political Mobilisation
Japan's rapid population ageing Á increasing life expectancy and declining fertility rate Á is a very well-known issue, currently shared by many other post-industrial societies, yet experienced by Japan several decades before them. Whilst in most developed countries elementary schools began emptying in the 1970s, in Japan the same event began in the 1950s (Peterson 2002). In such a scenario, the total fertility rate (TFR) went from an average of 4.42 during the baby boom (1947Á1949) 4 to an average TFR of 2.1 (replacement level) in 1960. In 1974 it decreased to 2.05 and continued on this downward trend until reaching a record low of 1.25 in 2005 (Itoh 2000; Statistics Bureau 2009). Simultaneously, the proportion of older people (65' ) increased rapidly, whereas the proportion of workingage people (15Á64 years) and younger population (0Á14 years) has been decreasing considerably since 1996 and 1982, respectively. By 2008, 22.1% of Japan's total population (127.6 millions) was 65 and older, 64.5% was of working-age and 13.4% was under 15 (Statistics Bureau 2009).
Obviously, these demographic transformations along with the increasing workforce participation of women led to a family-care crisis. Given the limited public and market welfare provision Japanese women started to criticise the family-centred welfare system, which clearly was not meeting their rising demands for elderly and child care (Eto 2001;Higuchi 2004). Despite women's social call, in 1981 the government introduced two directions in its social policy: (1) budget cutbacks in social services and (2) a campaign to revive the family and community as main sources of social welfare and care (Campbell 1992). Those most affected were middle-class housewives in their forties and fifties, who had formed the basis of the Japanese-style welfare society (1950Á1970s), and thus initiated a movement against the state's neoliberal policies. Amongst Japanese women's mobilisation are two groups that stand out by their achievements. First, the work of the Women's Association for a Better Ageing Society, initiated in 1982, is indeed significant. By addressing elderly care as the responsibility of the entire society, and not of individual families, a concept they referred to as ''socialised care'' (Eto 2001), this group helped redefine Japan's social policies. Moreover, it is through an active promotion of the issue of care, research surveys and tours to study public elderly care models in Europe and USA that this group managed to increase women's involvement in Japan's policymaking process (Higuchi 2004;Peng 2004). Another important group is the Kanagawa Club, created in 1971 as a cooperative society. Due to social and cultural problems affecting their daily lives, in 1985 this club established a non-profit welfare service to provide high-quality private services for elderly people in the Kanagawa Prefecture (Eto 2001).
These groups' activities and opinions on social care gained not only public support but respect from Japanese policymakers, particularly in the Ministry of Health and Welfare (MHW). Furthermore, they took advantage of the Liberal Democratic Party's electoral decline in the 1990s and of being invited to serve in the advisory council of the MHW in order to attain their demands. Indeed, as Peng (2001) and Eto (2001) claim, most of the social policy reforms in elderly care introduced in the 1990s, including the two most important, the Gold Plans and Long-Term Care Insurance, were only possible because of the philosophies of ''socialised care'' and ''social market model'' created by the Women's Association and the Kanagawa Club, respectively.

Education and Workforce Participation of Women
Nowadays, Japanese women are taking advantage of easier access to education. According to Ogawa and Retherford (1993: 588), over the past decades the educational attainment of women has been rapidly rising. In 1955, only 5% of women in the corresponding age groups were enrolled in junior college or university compared to 15% of men. By 1990, those figures had increased to 37% and 35%, respectively. In 2007, the enrolment in higher education represented 54% for women and 61% for men (UNESCO 2009).
As for participation levels in the workforce, the proportion of women aged 20Á49 in paid work increased from 13% to 42% between 1963 and 1990 (Ogawa & Retherford 1993: 588), and it rose steadily after the burst of the economic bubble in 1991 because companies began hiring more women as a cheaper source of labour. However, the majority were brought on as part-time or contract workers without benefits or job security, which is still the case as 90% of the eight million part-time workers in Japan are female (The Washington Post 2007). In contrast to men, who are the ultimate ''corporate warriors'', women are considered ephemeral and peripheral workers (Roberts 2005: 193). Certainly, the reason for this is that companies expect women to leave their jobs to take care of their small children 5 or frail relatives and fulfil the traditional feminine role that should match Japanese men's identity of an ''absent father/husband'' fully committed to his job. Thus, one could argue that these gendered working patterns disempower women in Japanese society, business and politics.
However, with women having greater access to higher education and paid work, the traditional gender division of labour, stressed by the ideal of the ''professional wife'' is fading away, albeit gradually. Japanese women, particularly those of younger generations, do experience a conflict between traditional societal expectations, lack of welfare support and their current individual opportunities and values, which altogether makes it difficult for them to express any discontent about their caregiving burdens (Hashizume 2000). Interestingly, though, the proportion of married women (aged 20Á49) ''working'' as family carers has steadily declined over the last 5 decades. For example, from 1963 to 1990 the ratio decreased from 13% to 10% (Ogawa & Retherford 1993: 588), which illustrates how women's increased workforce participation, as well as their increasing negotiation of and resistance to the traditional labour division must exert pressure on the Japanese Government. Nonetheless, this negotiation between traditional gender roles and individual opportunities is not limited to education and labour market, it also affects areas such as the distribution of household chores and the participation in family leisure activities, in which, as stated earlier, Japanese men are often absent.

Intergenerational Co-residence Patterns
As mentioned earlier, under the ie system the caregiving ideal was not only based on intergenerational co-residence, but on the assumption that it was the daughter-in-law's responsibility to perform all practical elder care duties in her husband's household. To date, however, Japanese women (i.e. daughters-in-law) are better educated and often work outside the home, and thus, as Long et al. (2009) note, it is more likely that they would negotiate living arrangements and traditional elder care expectations before and during marriage. Therefore, co-residence patterns have shifted with the decrease in three-generation households. Yet, when compared to other advanced nations, Japan still shows a high incidence of intergenerational co-residence. For instance, in 1975 three-generational households represented 18% of all Japan's households; by 2002 the proportion was reduced to 10%, whereas during the same period the share of one-person households increased from 18% to 23%. In contrast, households consisting only of members aged 65 and over, also grew in numbers representing 22% of all households by 2005 (Cabinet Office 2008; Ministry of Health, Labour and Welfare 2006). Considering these statistics, caregiving roles and expectations are being negotiated and, more importantly, seem to be shifting from the daughters-in-law towards spouses and adult children. 6

Marriage Trends
As Roberts (2005: 188) notes, amongst the reasons for Japanese women to postpone or even avoid marriage is the desire to continue working and enjoying a high living standard, away from the social burdens related to the traditional roles of ''wife'' and ''mother''.
Currently Japanese people are getting married at an average age older than in any other developed nation (Statistics Bureau 2009). For instance, in 1985 the mean ages at which women and men first married were 25.5 and 28.2 years, respectively, whereas for 2008 the mean age was 28.5 for women and 30.2 for men. As a result of these marriage patterns, motherhood trends are also impacted: the average mother's age on the birth of their first child has risen from 25.6 in 1970 to 29.5 in 2008 (Statistics Bureau 2009). Thus, this is an important factor behind the downward trend in Japan's fertility rates. The reason that a decline in marriage rates is correlated with the decline in fertility rates is because in Japan there is still a great social pressure that intends to protect the traditional family, and therefore discourages women to be single mothers, which visibly contrasts with other developed nations where it is more common to see children being born out of wedlock (Itoh 2000: 70). Furthermore, arguably, the reasons for the low-fertility rate are also rooted in the lack of child-care policies aiming at helping women balance work and home, as currently the government's emphasis is mostly on distributing child financial support, albeit relatively low in terms of GDP percentage. 7 Invariably, these conditions embedded in a male breadwinner model make Japanese women to reconsider fertility decisions. Thus, the burden of combining work and family life in addition to the high opportunity-costs associated with having a child are forcing women to choose between a career and marriage/family.
As we have seen, due to its demographic profile and various financial and social factors Japan is experiencing what Daly and Lewis (2000: 288) call a ''care crisis'', namely, a decrease in the supply of Á female Á unpaid caregiving labour, precisely at a time when the demand for caring is rising. In this case, the changes in individual attitudes and expectations of Japanese women regarding family care Á living arrangements, marriage and fertility Á as a result of a new set of opportunities are clearly impacting gender relations as well as Japan's social policy agenda. Thus, as Gottfried and O'Reilly (2002) claim, fertility rates do emphasise how individual decisions are affected by the gendered character of welfare regimes, in this case by a strong male breadwinner model, and how this, consequently, impacts welfare state development and social change.

Japanese State Responses
The shifts in gender relations and demographic patterns described above are clearly informed by a feminist discourse. Moreover, they seem to have created the ideal political and social context to set specific public policy goals and, above all, forced the state to redefine the issue of social care, as well as its own interests and direction regarding welfare provision. This section will examine how the Japanese Government is addressing the issue of social care by referring to two of the most recent social policy reforms: the policies in long-term care and promotion of a gender-equal society.

The Long-Term Care Insurance
Up until the 1980s, Japanese Government concerns were merely focused on the economic consequences of demographic ageing, such as social security problems associated with the increasing dependency ratio, savings rates decrease, workforce shortages and the consequent decline in productivity and economic growth, and ultimately, international competitiveness. Thus, unsurprisingly, as Peng (2004: 401) states, nearly 90% of Japan's social security expenditure was allocated to pension and health insurance systems. However, this outlook changed drastically in the 1990s. Due to fertility decline, public and policy debates shifted to the issues of women's employment, family-work imbalance, the declining marriage rate and the overall crisis of family-based care. Subsequently, from 1990 to 2000, social welfare expenditure doubled, most of which accounted for both elderly and child care (Peng 2004). 8 Despite this welfare expansion, Japan's public care system Á 1989s Gold Plan and 1994s New Gold Plan Á has been highly criticised and proven inadequate in meeting Japan's elderly care needs. In fact, in 1996, 85% of Japan's elderly people continued being cared for at home by their female kin (Peng 2002). It is clear, then, that under this ''expansion'' the Japanese Government was still relying on women as main care providers.
However, as a result of women's political mobilisation in 2000 the government introduced the Long-Term Care Insurance (LTCI) (Kaigo Hoken) under the slogan ''From Care by Family to Care by Society'' which clearly reflected not only women's lobbying actions but the overall societal tension between family-based care and state support. This system has made possible long-term care a universal right for every Japanese person aged 65 and over (or suffering from a disease derived from ageing) based only on physical and mental status, regardless of family circumstances and income (Long et al. 2009;Peng 2002). In this sense, as Campbell (2000) argues, LTCI symbolises a major expansion and a political shift in the development of Japanese social policy.
In line with Japan's long-standing tradition to adopt social policy from other countries, policymakers studied the European experience of social welfare systems and insurance programs. For instance, Tsutsui and Muramatsu (2005: 523) argue that the LTCI programme is less constrained by the national budget than tax-based models of expanding long-term care systems, such as in Scandinavian and UK, as well as in Japan's Gold Plans. Although designed in a similar way to the German LTCI programme (initiated in 1994), the Japanese system also incorporates Scandinavianstyle community-based management, in which municipalities, acting as insurers, manage their finances, set premium rates for people over the age of 65 and plan and administer services. Japan's LTCI also included an element of care management developed in the UK in which professional care managers (e.g. nurses, physicians, social workers, occupational therapists) prepare care plans for elderly people that may include schedules of home nurses, home helpers, day-care, equipment rentals, long-term residence and respite care (Long et al. 2009;Tsutsui & Muramatsu 2005).
Thus, by setting the notion of a ''social contract'', Japan's LTCI has established a market-oriented perspective in care services, aimed at achieving a greater mix of welfare providers with the main objective of relieving Japanese families of the care burden. Yet, to do so the government encourages the elderly to be cared for at home and is also promoting selfcare and disability-free independent life (Fuyuno 2007;Tsutsui & Muramatsu 2005) with the purpose of reducing reliance on costly longstay in hospitals and residential homes. However, the meshing of social security policy reform with a market-oriented approach has had other consequences as elderly people and their families are not necessarily receiving more or better care than prior to the implementation of LTCI. Inevitably, the combination of these factors has led both public and researchers to seriously question the quality and efficiency of social care services under Japan's welfare state expansion, along with deregulation and decentralisation processes (Peng 2001(Peng , 2002(Peng , 2004. Finally, what remains is to examine the effects of social care services provision and social policy reforms on gender issues. Under the current Japanese system most women have continued to be employed as low-wage, part-time and contract workers, and therefore, such social and economic disempowerment clearly adds more complexity to the redefinition of the boundaries of social care between the state, the family and the market.

Promotion of Gender Equality
Thus far, it has been established that the decrease in fertility is interlocked with changes in marriage trends, increased education and workforce participation of women, along with individual attitudes and expectations. However, at times, only the professional ''progress'' of women, and not the whole Japanese socio-cultural apparatus, has been held responsible for the recent socio-demographic trends. As Rostow (2000) notes, in 1996, even Japan's former Prime Minister, Ryutaro Hashimoto identified advancements in women's educational participation as the cause of population ageing, and proposed policies to discourage young women from pursuing a professional career. Such proposals, however, did not materialise due to the high public opposition. Instead, the Japanese Government soon realised how difficult it was for women to combine work, family-life and childbearing, and therefore initiated the Plan for Gender Equality 2000, which subsequently led to the Basic Law for a Gender-equal Society in 1999.
Indeed, ever since Japan's fertility rates have declined, the government has tried to establish a legal framework that will promote gender equality in both home and workplace (Equal Employment Opportunity Law of 1986 and 1989) and has also implemented several ''family-friendly'' policies (e.g. Angel Plans of 1995 and 2000; Child/Family Care Leave Law of 1991 and 1999; Law to Promote Support Structures for the Upbringing of the Next Generation of 2005) which aim at relieving women of the burden of childbearing, childrearing and elderly care while promoting the ''joint-participation'' (Osawa 2000) of men and women in household-related activities (Roberts 2005;Schad-Seifert 2006). However, only a very limited proportion of working women has benefited from these schemes. For example, Usui (2003: 20) explains that one out of three employed women is a part-time worker and approximately 20% of those are ''quasi part-time workers''; in other words, they work fulltime, however, because of their contractual status are excluded from the Child/Family Care Leave benefits. This, certainly, as Roberts (2005) notes, has forced working women into a dual role as there is a lack of men's involvement in household chores, let alone child and elderly care, which is evident in the average 48 min a day a Japanese man spends in helping with housework and child care (The Japan Times 2008). Therefore, it appears that all these social policies have the specific goal of enabling women to balance family and work responsibilities but less attention is given to men in the objectives of Japan's genderÁequality policies. Moreover, as Roberts (2005) and Schad-Seifert (2006) would argue, behind the government's measures for harmonising family-life and work is the objective to ensure women's workforce participation in order to revitalise Japanese economy and encourage working women to have more children. And although in recent years there has been a significant increase in the fertility rate, government policies should seriously address a redefinition of the traditional gendered division of labour, especially men's roles and work-oriented lifestyle, so that men can also be encouraged to balance work and family-life. This in turn could help ease the issue of family-care responsibilities as well as other pressures Japan's ageing society is currently facing.

Conclusion: The Shifting Social Á Care Á Contract
It has been demonstrated that gender relations and demographic trends have been key factors in Japan's most recent reforms in social policymaking. Until recently, the Japanese state has relied on women, families, communities and companies Á in that specific order Á to provide social care and welfare whilst giving priority to policies for economic and industrial development. Undoubtedly, the trajectory of change observed in the Japanese case is the legacy of a strong familialistic model, which creates new tensions and contradictions between societal shifts, changing gender roles and the existing welfare state arrangements. Thus, it is fair to say that a welfare system centred on the family ultimately produces a social and political discontent as members of the younger generations, particularly women, start seeing the tradition of family care and welfare responsibilities as a heavy burden. These issues are not particular to Japan: many other familialistic societies in Southern Europe (i.e. Italy, Portugal, Spain, Greece) and East Asia (i.e. Korea, Singapore, Taiwan and China) face similar socio-demographic trends (low-fertility rates/increased longevity) and changes in gender relations (Ogawa 2008;Peng 2002). Interestingly, due to historical and cultural similarities Á and a long tradition of economic development transfer Á East Asian countries are already viewing Japan as a model to develop their own ageing-related policies according to their particular needs, 9 especially since these societies appear to be ageing at a much faster pace than Japan (Ogawa 2008).
The Japanese welfare state makes, therefore, a strong case for leading the way into issues of population ageing, fertility decline and the accompanying imperatives for social care. As discussed earlier, Esping-Andersen's typology has identified Japan with the conservativeÁcorporatist regime but also highlighted its uniqueness as a hybrid model that cannot easily fit into the Western comparative framework. Moreover, the so-called uniqueness of Japan's welfare state may be understood in terms of its commitment to balance economic growth and demographic dynamics, which is clearly the premise for Japan's social welfare development. However, Esping-Andersen's model fails to fully explain the trajectories of change of the Japanese state, as it does not integrate further issues of gender relations into the analysis. Thus, by drawing on Daly and Lewis' (2000) model I have been particularly concerned to present the development of the Japanese welfare state from a gender-social care dimension. With gender being a crucial factor in the organisation of welfare, the use of a feminist/gender perspective was at place here. Furthermore, since much of the existing work on welfare states using this approach has been undertaken to compare Western regimes (Bergqvist et al. 1999;Korpi 2000;O'Connor et al. 1999) it appeared necessary to reflect on the Japanese case and recent feminist analyses by Ito Peng. For future studies it might be interesting to compare Japan and other East Asian countries in terms of gender regimes typologies in a similar way to Pascall and Kwak's (2005) study of Central and Eastern European countries. In their analysis they found that state policies in that region fail to support working women contradicting, therefore, the ideas and ideals of parental responsibilities. Women in these countries are similar to Japanese women in the sense that they are also negotiating a dual (work/ family) burden and expect care to be socially shared, especially between the family and the state. Thus, applying such type of analysis could bring new understandings about Japan, and above all could help us theorising the interplay of gender and welfare development in the East Asian region.
In conclusion, the case of Japan highlights three main issues. First, the trajectory of changes in the Japanese welfare state development shows us how social policies are undoubtedly gendered and come into being as a result of specific societal changes and pressures. In this context, the male breadwinner/professional housewife social contract has been undermined by the increase in education and work opportunities for Japanese women. Thus, the second issue is that such opportunities along with socioeconomic, political and demographic changes contributed to the political mobilisation of women who clearly succeeded to impact both the process and outcome of Japan's social welfare policies regarding the issue of social care, whilst also giving way to new individual views on gender roles to younger generations.
Although, today many Japanese women still uphold a strong sense of filial responsibility towards their elderly relatives, they are evidently negotiating and even rejecting the traditional feminine role inherent to the Japanese-style welfare state. This consequently leads us to the third main issue. Thus far, it seems that most of the social policy reforms have relieved women of the burdens of family care through an expansion of public provision of child and elderly care, albeit limited by neoliberal-residual principles. Nonetheless, it seems so far that Japan has not being able to recognise that the real problem behind low-fertility rate/population decline is not merely changes in women's roles but the necessity of gender equality in order to facilitate the balancing of work and family life for both men and women. Perhaps, one of the reasons for Japan's inability to change the basic structure of gender relations is rooted in the traditional ideal towards women, marriage and children. There is no denying, then, that the imperatives for change are embedded in a feminist discourse which aims to force the Japanese Government to carefully examine the structural issues behind societal and individual attitudes on social care and gender roles, in order to redefine more effective social welfare policies for its ageing society.