Introduction

1Eldercare is an important economic, social, and health concern in ageing societies. Globally, Japan has the highest number of elderly people in proportion to its total population. The need for long-term care has therefore been growing rapidly, but this increasing demand sometimes remains unsatisfied, leaving elderly people in painful situations. Since the government has a moral obligation to ensure a minimum level of eldercare, it has been doing so within the existing framework. However, the outbreak of the Covid-19 pandemic (hereafter, “health pandemic”) has complicated the provision of eldercare. Many caregiving institutions have either shut down or substantially reduced their services, thereby greatly increasing the demand for family caregiving. Due to the social distancing rules and other increased health safety measures, the provision of family caregiving has become challenging. The health pandemic has forced the redesign of eldercare services; the modes of services must be considered from a new perspective.

2In an ageing society, increasing numbers of elderly people require assistance in their daily lives. This is particularly imperative for those who have chronic illness and disabilities. Family caregiving is the act of assisting elderly family members with daily life activities, health care, social and financial issues, among others. It has traditionally been perceived as an act of kindness, love, and loyalty to the elderly family members needing care. As a result, family caregivers hold the primary responsibility to provide eldercare, and comprise the largest and most trusted source of eldercare in society (Lee, 2004; Redfoot, Feinberg & Houser, 2013; Ting & Woo, 2009). However, the changing family structure, more elderly people living alone and apart from families, greater female participation in the labor market, higher socio-economic status, and fewer eligible caregivers in the family have highlighted the need for formal eldercare (de Meijer et al., 2011; Ku, Liu & Wen, 2013; Li et al., 2017; Rogero-Garcia, 2009). Donald Redfoot et al. (2013) provided evidence for a decreasing trend in family caregivers that could continue in the future. The stress associated with caregiving has increased for existing caregivers because of this decreasing trend. Previous studies have provided evidence on the difficult, burdensome, and stressful nature of the job, which sometimes causes physical and mental problems to the family caregivers (Okamoto, Hasebe & Harasawa, 2007; Pinquart & Sorensen, 2007; Schulz & Sherwood, 2008). On the one hand, family members choose to provide eldercare because of its ties to cultural and family values; on the other hand, the stressful nature of the job limits their capacity to do so. Combined with the changing family structure and recent health pandemic, the family caregiving issue has taken a multidimensional appearance. Existing care market models suggest that the demand for care services because of reduced family caregiving and limited capacity of institutional caregiving must be solved by the existing market mechanism (Kadoya, 2018). In order to achieve this, it is important to understand the attitude toward family caregiving and its determinants.

3Japan is one of the fastest ageing countries in the world. Recent studies have shown that 28.4% of people are aged 65 years or older, and this number is expected to reach 40% in 2040-2050 (Yamada & Arai, 2020). The number of older people with a disability was 4.71 million in 2019. The number of older people living alone was 5.93 million in 2015, and is expected to increase dramatically in the near future. Thus, the provision of eldercare for a large number of older people will be a big challenge for a country like Japan, where the portion of the active population (aged between 40 and 64 years) has been reducing. Like the global phenomenon, eldercare in Japanese society has a long tradition, and is guided by socio-cultural norms and traditional values (Asai & Kameoka, 2005; Yamaguchi, Cohen & Uza, 2016; Young, McCormick & Vitaliano, 2002). In the current culture, elderly people hold respectable positions in the family and, thus, are accustomed to receiving care from family members. Although adult children are the primary source of family eldercare in Japan, care services vary substantially with gender, marital status, and the children’s opportunity cost (Hanaoka & Norton, 2008). Traditionally, females are expected to assume the responsibility of eldercare in Japanese society. However, the likelihood of becoming a primary caregiver is higher for single daughters with lower education (Tokunaga & Hashimoto, 2017; Tokunaga, Hashimoto & Tamiya, 2015). In addition to cultural and traditional reasons, expected bequests from parents appear to be an incentive for providing family care (Horioka et al., 2018). However, Japanese parents are less likely to leave bequest to their children unless they receive intergenerational transfers from their parents or parents-in-law (Horioka, 2014, 2009; Niimi & Horioka, 2018). Moreover, the introduction of long-term care insurance (LTCI) added a new dimension to the provision of eldercare in Japan. Japan introduced LTCI in 2000 in the context of changing population demographics and the increased need for long-term care. The LTCI in Japan is similar to the social insurance system available in Germany, the Netherlands, Luxembourg, and Korea (Miyawaki et al. 2020). All people above 40 years of age are required to participate and contribute to LTCI. Every individual above 65 years of age (or above 40 years of age with particular diseases) are entitled to receive LTCI services at the institutional, home, and community level. Under the provision of LTCI, recipients are required to pay for 10% of the services, whereas the remaining 90% are paid from the LTCI budget (Yamada & Arai, 2020). As a result, people now have an incentive to seek formal eldercare, such as homecare, daycare, short-term care services, among others (Sugawara & Nakamura, 2014). Moreover, Shinya Sugawara and Jiro Nakamura (2014) provided evidence for the positive impact of the introduction and diffusion of LTCI on female labor supply, i.e., females being more available for formal employment. However, even after the introduction of LTCI, adult children continue to comprise the largest supply of family eldercare in Japan (Hanaoka & Norton, 2008). Several studies have provided evidence on family caregivers’ challenges in providing eldercare. Hiroyuki Yamada and Satoshi Shimizutani (2015) studied the labor market effect of informal family eldercare and found that family care provision adversely affected the probability of working, employment status, and hours worked, particularly for females. Yoshinori Nishimura and Masato Oikawa (2017) found that informal, family eldercare had a negative influence on female employment, but did not influence male employment. However, Takashi Oshio and Emiko Usui (2017) argued that informal parental care had little effect on female employment after controlling for endogeneity and time-invariant heterogeneity. Takashi Oshio (2014, 2015) found that middle-aged adults experienced distress when providing family eldercare. The magnitude became even higher for female caregivers in special circumstances. Yoko Sugihara et al. (2004) provided evidence for the emotional exhaustion and depression of caregivers. While emotional exhaustion was found to vary among individuals, depression showed little variation. Female caregivers were found to face more exhaustion than others. Previous studies also provide evidence on the effective management of stress and burden to provide family care with relative ease. Sayoko Uemura et al. (2014) studied Japanese male family caregivers in terms of burden, stress, and well-being. They found that male caregivers used nursing-care services to alleviate the stress and burden of family care, which helped them be able to enjoy leisure time more often.

4Thus, we observe two distinct and competing forces that shape family caregiving in Japan. While family members have been providing eldercare because of cultural and family values, the incentives derived from the LTCI, changing family structure, and greater female labor market participation have been forcing them toward formal care services. It is thus important to conduct a comprehensive study on the attitudes toward family caregiving to better understand the supply of family caregiving in the future. The Preference Parameters Study (PPS) of Osaka University, Japan has provided an opportunity to study family caregiving to understand the forces behind the formation of these attitudes in Japan. This study examines how socio-economic factors are related to negative attitudes toward family caregiving. We argue that people who have a negative attitude toward informal family caregiving would potentially seek formal caregiving. Thus, understanding the socio-economic factors that are responsible for negative attitudes toward informal family caregiving would also help to understand the potential demand and supply of formal caregiving. To the best of our knowledge, this is the first study to investigate how socio-economic factors are related to attitudes toward family caregiving. Our study makes a significant contribution to the existing literature in the understanding of the reasons behind family members losing interest in providing informal caregiving and, instead, being inclined to rely upon formal care providers.

Data and methodology

Empirical findings

11The five models differ in the number of explanatory variables used to explain attitudes. Organizing the models in this way helps to understand the consistency of results when new variables are added to the models. Model 5 is the most comprehensive model, which includes all explanatory variables.

12Table 4 shows the probit regression coefficients. Female respondents are more likely to have negative attitudes toward family caregiving. The result is very strong and consistent across the models, indicating that the effect does not disappear even after controlling for more variables. Respondents’ age is positively related and age squared is negatively related, indicating that a non-linear positive relationship exists between age and negative attitudes toward family caregiving. However, the evidence is found only in Model 1, and disappears when more variables are added to the models. We find a negative relationship between being employed and family caregiving attitude, meaning that unemployed respondents are more likely to have negative attitudes toward family caregiving. However, the relationship is not consistent because the effect disappears in Models 4 and 5. Married respondents are more likely to have negative attitudes toward family caregiving. This relationship is consistent across the models. We further find that the more sons the respondents have, the more likely they are to have negative attitudes toward family caregiving. The results are consistent across the models. Respondents’ annual household income is negatively related to negative attitude, indicating that respondents with lower income are more likely to have negative attitudes toward family caregiving. Although the relationship is not significant in Model 4, it is significant in Model 5, which is the most comprehensive model. Finally, respondents who believe that the government should be responsible for financially dependent people are less likely to have negative attitudes toward family caregiving. These findings were consistent across models.

13However, we find that the relationships between negative attitude toward family caregiving and education, number of siblings, number of daughters, household financial assets, proportion of income after retirement coming from social security, savings for living expenses and long-term care, inheritance or monetary transfer, parents’ education, and whether one parent or both parents require care are not significant.

14To check the robustness of the results and to confirm that they do not contain a selection bias, we estimated all models using the same subsample as used in Model 5. The signs and significance of the variables remain the same. Our results are robust and do not have a selection bias. We do not report the re-estimation results to save space, yet they are available upon request.

Discussion

15This study investigated socio-economic background of the family caregiving attitude in Japan. Our results show that females are more likely to have negative attitudes toward family caregiving. Females have traditionally played a major role in family caregiving (Bookman & Kimbrel, 2011; Gupta, 2009; Huntsberry-Lett, 2020; Tokunaga & Hashimoto, 2017). Thus, reasons for females being reluctant to provide family caregiving should be studied in order to better understand the potential supply of family caregivers. There could be several reasons for these negative attitudes toward family caregiving. In the current context, females have been engaging in the labor market and undertaking formal employment more than before, leaving less time available for family caregiving (Bianchera & Arber, 2007; He & McHenry, 2016). These females are likely to be under immense stress to provide eldercare with the limited available time, making them susceptible to have negative attitudes toward family caregiving. As a result, the females providing eldercare have been facing more stress and developing mental conditions than others (Oshio, 2015; Sugihara et al., 2004). Consequently, females, compared to males, favor institutional formal care over informal family care (Carvalho et al., 2019). Our results further show that married people are more likely to have negative attitudes toward family caregiving. Previous studies provided evidence that married people assume the responsibility of eldercare (Kadoya & Khan, 2017a, b). We argue that the burden and stress associated with the perceived responsibility for eldercare lead to a negative attitude toward family caregiving. Since married people include married women who have traditionally played the main role in family caregiving, the stress and burden associated with providing eldercare could be the reason why married women have become uninterested in family caregiving. It has been evident that married women in Japan hold the primary responsibility of providing eldercare to their parents-in-law, and feel additional pressure and stress when providing care to them (Kadoya & Khan, 2017a, b; Oshio, 2014; Wakabayashi & Kureishi, 2018). To examine the conjecture that married women, who are responsible for taking care of their parent-in-laws, have negative attitudes toward family caregiving, we re-estimated our models using an interaction term between the variables female and married. Table 5 shows that the regression coefficient of the interaction between gender and marital status is positive and significant. Moreover, the significance of the female variable disappears when the interaction term is used. The results indicate that married women who are presumably responsible for taking care of parent-in-laws have a negative attitude toward family caregiving.

16Another demographic issue that shapes respondents’ family caregiving attitude is the number of sons they have. Respondents are more likely to have negative attitudes toward family caregiving the more sons they have. Although sons and their families traditionally have caregiving responsibility (Kadoya & Khan, 2017a, b; Kadoya & Yin, 2014), our results suggest that parents have a more negative attitude toward receiving eldercare from their children, the more sons they have. Our results somewhat contradict the traditional belief but are not entirely unsupported. Previous studies show that Japanese people are traditionally not reliant on their sons (Horioka, 2019), instead making them more reliant on formal care. Consequently, people develop negative attitude toward receiving and providing family care.

17Respondents with a lower household income are more likely to be pessimistic about family caregiving. Previous studies found that people in relatively low income groups could not afford the desired level of formal care (Bruhn & Rebach, 2014; Schmidt, 2017), and that depressive symptoms were prevalent (André et al., 2014). The pressure, stress, and depression associated with family caregiving, along with the lack of capacity to avail formal care, are likely to make people in relatively low income groups pessimistic toward family caregiving. Moreover, people’s perceptions of whether the government should support financially dependent people also shape family caregiving attitudes. We argue that the association between respondents’ perceptions of government support to financially dependent people and family caregiving attitudes depends on how formal and informal caregiving are associated. Respondents who are in favor of government support might prefer less participation by informal caregivers. Thus, they might have a negative attitude toward family caregiving. This line of argument is consistent with the view that the availability of formal caregiving reduces informal caregivers’ participation. The findings of Atsushi Miyawaki et al. (2020) are consistent with this argument. On the other hand, respondents who are in favor of government support might have a positive attitude toward family caregiving as well, expecting that government intervention would increase capacity to higher formal care services and, eventually, would reduce the burden and stress associated with informal caregiving. This line of argument is consistent with the view that the availability of formal caregiving might complement informal caregiving. Previous studies also provide evidence that the availability of formal caregiving and public intervention sometimes complements informal caregiving (Bonsang, 2009; Bolin et al., 2008; Denton, 2010; Saraceno & Keck, 2010). Our results show that respondents who perceive that it is the government’s responsibility to take care of financially dependent people were also found to hold positive attitudes toward family caregiving. These results are consistent with the complementary view of formal and informal caregiving.

18Although previous studies provide evidence that full-time employment makes people less available for family caregiving and more inclined to formal care, particularly for females, we found weak and inconsistent evidence that unemployed respondents have negative attitudes toward family caregiving. However, the effect becomes non-significant when the social security, perception of government support, and savings behavior variables were added to the model. Similarly, we found limited support for the evidence that age influences attitudes toward family caregiving.

19In light of the empirical evidence, we found that several socio-economic issues have driven people to have negative attitudes toward family caregiving. Overall, the socio-economic issues indicate that changing family structure, the pressure and stress related to caregiving, and having a negative attitude toward government support lead people to have negative attitudes toward family caregiving. Consequently, this attitude puts pressure on the provision of formal care, which is still insufficient in the country. Since the government is obligated to ensure a minimum level of eldercare, it must find ways to fill this gap. OECD countries have already started to provide eldercare through competitive markets in response to increasing needs (Kadoya, 2018). Historically, in Japan, public institutions started to intervene in the care market since the enactment of the Long-Term Care for Older People Law (Rojin fukushi hou) of 1963. Moreover, since the late 1980s, eldercare services from the private sector have gradually increased. The LTCI Law of 2000 deregulated private sector access to the market. Despite all these efforts, the decreasing trend in family caregiving means that formal institutional support cannot fulfill the demand for eldercare. This sector has been facing a shortage of qualified professional caregivers. Moreover, the emergence of the current health pandemic has not only reduced the existing institutional supply of eldercare but also made the provision of formal and informal family caregiving much more complicated. The potential reduction in the supply of informal family caregiving, as our study suggests, would further complicate the situation. The care market model suggests solving the issue through the market mechanism (Kadoya, 2018). The government had been contemplating outsourcing caregivers from other countries. We argue that this problem should be solved comprehensively within the existing market framework, taking both formal and informal family caregiving into consideration. There are several ways through which the supply of formal and informal caregiving can be increased. First, the supply of institutional caregiving could be increased. Previous studies found that the relative low wages and other benefits disincentivize prospective caregivers from joining the market (Farrell, 2018). By providing training, increasing salaries and job-related benefits, and improving working conditions, formal caregivers’ efficiency can be improved. Moreover, the introduction of artificial intelligence and robotics in the care market can reduce the burden on caregivers. Utilization of technologies would help retain the social distancing rules as well.

20Second, the supply of informal caregiving could be increased by providing cash benefits to family caregivers in the public long-term care insurance system. The cash benefits would compensate for their opportunity costs and foster a positive attitude toward family caregiving. Third, expansion of the existing nursing care leave system, where workers are allowed to take temporary leave to provide care services to elderly family members, could also be helpful in solving the problem. Expansion of the nursing care leave provision will presumably help children in complementing formal family caregiving and increase the supply of informal family caregiving.

Conclusion

21In an ageing country, due to tradition and culture, family caregiving plays the most significant role in eldercare. However, family caregiving has been facing challenges because of changing family structure and socio-economic issues. The outbreak of Covid-19 has made the provision of family caregiving even more challenging than before. As a result, demand for formal caregiving has been rising sharply, making it difficult to supply in the existing framework. Given this situation, it is important to investigate why people have negative attitudes toward informal, family caregiving and are more inclined to formal care. The objective of this study was to investigate how socio-economic factors are related to attitudes toward family caregiving. Using data from the PPS of Osaka University, Japan, we used several probit regression models to investigate this matter. Our study shows that a significant number of respondents have negative attitudes toward family caregiving. The regression coefficients show that gender, marital status, number of sons, household income, and perception of government support have a strong and consistent relationship with attitudes toward family caregiving. Respondents who are female, married, have relatively more sons, have a relatively low income, and are not in favor of government support to financially dependent people are more likely to have negative attitudes toward family caregiving. Although not strong and consistent across the models, a negative relationship with employment status is also evident. Previous studies on the provision of eldercare show that caregivers are generally females, married, and have more sons, more daughters, and more siblings (Huntsberry-Lett, 2020; Tokunaga & Hashimoto, 2017; Kadoya & Khan, 2017a, b). However, the results of our study show that family members who are traditionally perceived as being potential caregivers have negative attitudes toward family caregiving.

22This trend conveys an important implication for the potential supply of family caregiving, which policymakers should consider when framing ageing policies. While the government is contemplating outsourcing caregivers from outside Japan, we argue that the solution lies in enhancing existing caregivers’ efficiency through increasing salaries, improving working conditions, providing training, and incorporating technologies in the eldercare market. Moreover, ensuring informal caregiving during this health pandemic is needed to reduce the pressure on formal caregiving.

23The spectrum of family caregiving requires further investigation in the future. While we explored socio-economic factors responsible for negative attitudes toward family caregiving, we need to know the socio-economic background of the family caregivers who have positive attitudes toward family caregiving. We should also separate negative care giving and negative care receiving attitudes. The comparative analysis would help us to better understand the family caregiving issue and potential demand for formal caregivers. If family caregiving could be retained by policy intervention, the pressure on formal care would be reduced and elderly people who prefer to receive care from family members would be satisfied.