Introduction

Studies of defamilisation/familisation and adult worker models have been steadily growing (Reference AnnesleyAnnesley, 2007; Reference DalyDaly, 2011; Giullari and Lewis, Reference Giullari and Lewis2006; Kilkey and Merla, Reference Kilkey and Merla2014). These studies can to a certain extent be seen as a response to the decline of the male breadwinner model. With the focus on the link between these studies, this paper is intended to bring theoretical advancement concerning the examination of the adult worker models. It particularly focuses on demonstrating how the discussion of the government pro-employment and decommodification measures for reducing the defamilisation and familisation risks faced by women contributes to the examination of the adult worker models. It has three objectives. The first is to categorise the adult worker model into four types (market-focused, supported, choice-focused and collective consumption) based on different combinations of the pro-employment and decommodification measures. The second is to explore the relative desirability of these four types in enhancing women's well-being. The third is to discuss issues concerning the application of the adult worker models in the analysis of the government responses to defamilisation and familisation risks.

The paper starts by discussing the concepts of defamilisation and familisation risks. This is followed by the analysis of the role of the government pro-employment and decommodification measures in assisting women to tackle defamilisation and familisation risks. In the third part, the similarities and differences between the four types of the adult worker model in terms of different pro-employment and decommodification measures are explored, and the relative desirability of the four types in enhancing women's well-being is discussed.

In the fourth part, we discuss the empirical example of the pro-employment and decommodification measures provided by the Hong Kong government. By doing so, we study issues concerning the application of the adult worker model in the analysis of the policy responses to the defamilisation and familisation risks in the empirical world. Our focus is on the gap between the pure form and the actual form of the adult worker models. The four types of the adult worker model presented in this paper are by nature ideal typical. The actual forms of the adult worker models upheld by the government may vary from them. As shown in the later parts of this paper, the type of the adult worker model associated with the Hong Kong pro-employment and decommodification measures does not fit totally in any of the four pure types. Instead it is linked partially to more than one types. Hence, discussing the empirical examples of the pro-employment and decommodification measures in Hong Kong makes us more aware that the actual form of the adult worker model that a government develops is not necessarily fully in line with the pure form of the four types of the adult worker models discussed in the paper.

Defamilisation/familisation risks

Esping-Andersen (Reference Esping-Andersen1990) has famously classified 18 OECD countries into a typology primarily based on the concept of ‘decommodification’, which has commonly been understood as the extent to which individuals can maintain a socially acceptable standard of living regardless of their market performance (Powell and Barrientos, Reference Powell and Barrientos2011). The focus on ‘decommodification’ has thereby primarily been concerned with how states respond to common labour market risks, which manifest in the absence of one or both of two conditions that affect how people organise their own welfare (Kilkey and Merla, Reference Kilkey and Merla2014; Reference OrloffOrloff, 1993): the freedom to choose whether to take part in the work economy, and the opportunity to have a reasonable standard of living. Without both of these conditions being met, individuals may have to live in poverty, unless they are prepared to take part in the job market irrespective of working conditions. By enabling individuals to secure a reasonable standard of living outside of formal employment through the provision of welfare policy, it is arguably easier to tackle labour market risks.

The comparative social policy literature has been heavily influenced by Esping-Andersen's (Reference Esping-Andersen1990) distinction of ‘Three Worlds of Welfare Capitalism’ and its various critiques and additions (Hudson and Kühner, Reference Hudson and Kühner2009; Reference KrogerKroger, 2011; Ku and Finer, Reference Ku and Finer2007). Indeed, Esping-Andersen (Reference Esping-Andersen1990) has famously been criticised for being insensitive to the risk faced by women in dealing with family issues (Kilkey and Merla, Reference Kilkey and Merla2014). To many women, it is not dependency on the labour market but the unequal gender division of unpaid responsibility that undermines their welfare (Reference BambraBambra, 2007). Moreover, such an (largely) involuntary gender division of labour often limits women's choices and possibilities to develop their career (Reference KrogerKroger, 2011). Furthermore, not all women prefer to play the role of full-time carer; some may want to take an active part in the work economy as a worker so as to achieve financial autonomy in the family (Keck and Saraceno, Reference Keck and Saraceno2012; Reference ListerLister, 1997). Hence only supporting women to lead a decommodified life may not necessarily be enough to meet women's diverse preferences.

In responding to Esping-Andersen's work (Reference Esping-Andersen1990), there has been a rising number of studies on the two concepts of ‘defamilisation’ and ‘familisation’ (Reference DalyDaly, 2011; Keck and Saraceno, Reference Keck and Saraceno2012; Reference KrogerKroger, 2011; Lohmann and Zagel, Reference Lohmann and Zagel2016). While focusing on the family rather than on market relationships, the defamilisation and familisation literature is thereby similarly concerned with individuals’ freedom from compulsion to enter into potentially oppressive relationships. For instance, in discussing the concept of defamilisation, Lister (Reference Lister, Baldwin and Falkingham1994) stresses the importance of finding ways to assist individual adults to uphold a socially acceptable standard of living independently of family relationship. Some studies of defamilisation focus on the terms and conditions under which women can choose to engage or not engage in caring relationships. It has been pointed out that some women (and men) may prefer to see their life anchored in the family rather than seeking a high degree of financial autonomy from family relationships (Reference DalyDaly, 2011). In other words, they may seek to play a more active role in the family as a provider of care and/or prefer to receive care from close family members (Keck and Saraceno, Reference Keck and Saraceno2012; Lohmann and Zagel, Reference Lohmann and Zagel2016).

The above studies provide important insights into the nature of defamilisation and familisation risks. Similar to more general labour market risks, defamilisation and familisation risks can be said to manifest in the absence of one or both of two conditions that directly affect how particularly women organise their welfare: women's opportunities to secure a reasonable standard of living, and women's freedom to choose whether and how to perform different roles (such as care provider and worker) within and outside the family.

The following are examples found in the literature of common defamilisation and familisation risks faced by women:

1. (a) Some women may rely financially on male family members and as a result have to perform the role of care provider in the family involuntarily (Reference NybergNyberg, 2002).

2. (b) Some women may want to take part in the paid labour market and perform the role of the financial provider but fail to do so because they are required to look after their family members as a full-time family carer (Reference BambraBambra, 2007).

3. (c) Some women may want to perform the role of a full-time family carer but fail to do so because they are required to earn their living independently in the paid labour market (Reference SaracenoSaraceno, 2015).

4. (d) Some women may be required to take up most of the caring responsibilities in the family involuntarily, despite their active participation in the paid labour market (Giullari and Lewis, Reference Giullari and Lewis2006).

In view of these examples, it is clear that the defamilisation and familisation risks are associated with two problems, which undermine women's well-being: gender inequality and women's lack of sufficient freedom to choose their way of participating (or not participating) in the family. Gender inequality could therefore be manifested in the unequal allocation of paid work and unpaid work between women and men. Orloff (Reference Orloff2009) has written that providing care is the source of many of women's economic and political disadvantages in a wage economy. Hence, eliminating the gender division of labour not only in paid work but also in unpaid work is often seen as an important means for promoting women's access to welfare (Reference KurowskaKurowska, 2018). For this reason, analysts have drawn attention to the dual-earner/dual-caregiver model (Reference CromptonCrompton, 1999; Reference Pfau-EffingerPfau-Effinger, 2005).

Women's participation in the family is concerned with the delivery of several roles such as the care provider, receiver of financial support and provider of financial support. Their lack of freedom to choose the roles they prefer can be caused by a lack of sufficient resources and/or the existence of constraints, or even by both a lack of resources and the presence of constraints.

There are different views on the relationship between these two problems. Some studies especially those informed by the capability approach perspective point out that even if women have the opportunity to choose the ways of participating (or not participating) in the family, they may not necessarily make use of this opportunity to reduce the gender inequality or advance their personal interests (Reference KurowskaKurowska, 2018; Lewis and Giullari, Reference Lewis and Giullari2005). To illustrate this point, Kurowska (Reference Kurowska2018) particularly highlights two different reasons why women act as a full time carer in the family instead of a worker in the work economy:

1. (a) A woman who stays at home and care for her child because she does not have a choice; and

2. (b) A woman who stays at home and care for her child (instead of pursuing employment) because she consciously chooses to – even if it is possible for her to take part in the work economy.

The presentation of these two cases suggests the importance of avoiding the presupposition that giving women the freedom to choose the roles they prefer to perform inside and outside the family would necessarily result in gender equality in the allocation of paid work and unpaid work.

However, some analysts argue that individuals’ agency freedom can be restricted along gender lines (Lewis and Giullari, Reference Lewis and Giullari2005). It is important to note that women do not necessarily make decisions concerning whether and how to participate in the family on a clean plate; instead this kind of choice is made in the context of gendered inequalities in power relations (Reference CromptonCrompton, 1999). In order to provide a more favourable condition for women to make genuine choices concerning the lifestyle they value, it may therefore be necessary to secure a more equal allocation of unpaid work. One widely discussed way of doing so is to encourage males to spend more time on unpaid work in the family through the provision of paid ‘daddy leave’ (Lewis and Giullari, Reference Lewis and Giullari2005; Reference SaxonbergSaxonberg, 2013). To date, the impact of this type of leave on the gender division of labour in the family is subject to debate. Some countries such as Japan and Korea show that the take up rate of the paid daddy leave is low. Analysts explain this problem by stressing that the cultural values which favour the male breadwinner model affects the effectiveness of the daddy leave as a tool for reducing gender equality (Chau et al., Reference Chau, Foster and Yu2017). However, studies show that fathers’ use of leave can have long-lasting effects on their subsequent involvement in the provision of care for their children (Haas and Hwang, Reference Hass and Hwang2008). Moreover, it has been discovered that the leave regulations create norms about good motherhood and fatherhood, and in turn transform to a certain extent the social construction of ‘normal’ gender roles (Reference LeiraLeira, 2002).

The discussion of different views on the relationship between gender inequality and women's lack of sufficient freedom to choose the ways of participating in the family has two implications on the search for ways to tackle defamilisation and familisation risks. First of all, the impacts on both of these problems should be considered when examining the effectiveness of the government's measures for reducing the defamilisation and familisation risks. Moreover, it is necessary to consider encouraging men, by using sticks and carrots, to take on more unpaid work as an important way for reducing the defamilisation and familisation risks faced by women.

Decommodification measures and pro-employment measures

How women respond to defamilisation and familisation risks can be highly related to how they respond to the paid labour market. As mentioned above, in order to reduce their financial dependency on other family members and increase their bargaining power over the allocation of caring responsibilities within the family, women may prefer to earn money by taking part in the paid labour market rather than playing the role of full-time family carer. Hence, it is possible that those government measures that assist women to participate in the paid labour market may at the same time increase women's freedom to choose whether and how to take part in the family. To illustrate this point, two kinds of government measures to reduce women's defamilisation and familisation risks can be separated – ‘decommodification measures’ and ‘pro-employment measures’:

1. (1) Decommodification measures refer to any government measures (social policies and regulations) intended to assist women to maintain a socially acceptable standard of living without taking part in the paid labour market.

2. (2) Pro-employment measures refer to any government measures (social policies and regulations) that assist women to maintain a socially acceptable standard of living in their working life through taking part in the paid labour market. They consist of:

   1. (a) Making work pay measures which are designed to make paid employment financially more attractive than remaining on benefits or dependent on a male-breadwinner (Reference AnnesleyAnnesley, 2007); and

   2. (b) Condition building measures which are designed to create favourable conditions for women's participation in the paid labour market (Reference BambraBambra, 2007). Unlike making work pay measures, condition building measures do not exclusively serve those who are already in the paid labour market.

Decommodification and pro-employment measures have the potential to assist women to tackle the above defamilisation/familisation risks by allowing them to play the role they prefer inside and outside the family and at the same time maintain a reasonable standard of living:

1. (a) By ensuring that women are eligible for state benefits, such as family, child and care allowances or unemployment benefits, the government can assist women to seek financial autonomy in the family (as a kind of decommodification measure). By ensuring that women are eligible for public services, such as early education, child and elderly care, the government can reduce women's caring responsibilities and give women more time to develop their career (as a kind of condition building measure).

2. (b) By giving female workers extra financial resources, through statutory minimum wage policy, paid maternity leave, and job-related retirement programmes, the government can assist women to create a better condition for them to have a reasonable standard of living (as a kind of making work pay measure). Female workers may be able to make use of these extra financial resources to purchase private caring services to reduce their caring responsibilities.

As mentioned above, women may have different preferences in relation to the role they perform in the family. Hence, women do not necessarily face the same kind of defamilisation risks or familisation risks. In order to meet the diverse needs of women, it is therefore necessary for the government to provide both decommodification measures and pro-employment measures (see Table 1). As also mentioned above, in order to reduce gender inequality, it is necessary to provide measures that encourage men to take up more unpaid work in the family. Examples of these measures are the decommodification measures that enable men to maintain a reasonable standard of living without taking part in the paid labour market and condition building measures (such as ‘daddy leave’). However, we should not presuppose that governments are always keen to increase women's freedom to choose the ways of participating (or not participating) in the family or to reduce gender inequality. In addressing the familisation and defamilisation risks, governments may have other considerations in mind, such as for example, to what extent these policy measures would intervene in the local economy, and strengthen the incentive to work. To illustrate this point, the relationship between decommodification and pro-employment measures and how they foster different adult worker models is discussed in the next section.

Adult worker models

Unlike the male breadwinner model, the adult worker model emphasises the importance of assisting not only male adults but also female adults to respond to changes in the global knowledge economy and participate in formal employment. Lewis and Giullari (Reference Lewis and Giullari2005) have compared the approaches to promoting women's participation in the paid labour market in the United States and the Scandinavian countries. The US government has traditionally provided a minimum level of welfare so that women (and men) have no choice but to maintain their standard of living through selling their labour in the work economy (Reference LewisLewis, 2001). In contrast, the approach used by the governments in the Scandinavian countries stresses the provision of care services for children and the elderly. Crucially, this approach treats women as ‘workers’ but makes allowance for difference by designing transfers and services in respect of care work for partnered and un-partnered mothers alike (Lewis and Giullari, Reference Lewis and Giullari2005). With reference to this contribution by Lewis and Giullari (Reference Lewis and Giullari2005), the authors in their previous work have distinguished three kinds of adult worker models – market-focused, supported and choice-focused (Yu et al., Reference Yu, Foster, Chau and Lee2017), whereby:

1. (a) The market-focused adult worker model is indebted to the US approach to women's participation in the paid labour market. As such, governments in favour of the market-focused adult worker model are likely to keep state benefits at a low or minimum level, giving women little choice but to earn their living through taking part in the paid labour market if they want to seek financial autonomy in the family.

2. (b) The supported adult worker model emphasises a generous provision of services in relation to the care of family dependents and cash transfers in respect of parental leave (Giullari and Lewis, Reference Giullari and Lewis2006; Reference LeitnerLeitner, 2003). With the support of these services, women are expected to have their caring responsibilities in the family reduced thus enabling them to spend more time in paid work.

3. (c) The choice-focused adult worker model stresses the importance of respecting women's rights including the right to work and the right to access welfare. This approach also implies that a woman should not only be regarded as a worker in the paid labour market, but also as a citizen who has the right to choose her own way of life. For instance, Saraceno (Reference Saraceno2015) has argued that women's activity in the private/unpaid sphere should be regarded just as valuable as the paid work in the public sphere. Following this logic, women should be given the support to choose between different options including the role of the main carer in the family; taking part in the paid labour market on a full-time basis; and providing care in the family and taking part in paid work at the same time.

Different adult worker models are associated with different ways of providing decommodification and pro-employment measures (see Table 2). If the government supports the market-focused adult worker model, it is likely to keep the provision of both decommodification and pro-employment measures to a minimum. The supported adult worker model stresses women's participation in the paid labour market and is likely to devote much more resources to the provision of pro-employment measures (making work pay and condition building) than to decommodification measures. Moreover, to further create favourable conditions for women's participation in the paid work economy, the government may encourage men to do more unpaid work through the provision of decommodification and condition building measures for men. If the government upholds the choice-focused adult worker model, it should provide a mix of the decommodification and pro-employment measures (making work pay and condition building) to meet the different preferences of different women on how to participate in the family and the paid labour market. As with the supported adult worker model, the choice-focused adult worker model can also be strengthened through the provision of decommodification and condition building measures for men. Moreover, by doing so, developing the choice-focused adult model may also serve to strengthen the dual-earner/dual-carer model.Footnote ¹ This paper suggests the collective consumption type as the fourth adult worker model. As with the choice-focused adult worker model, this model emphasises the importance of recognising and respecting women's status as a citizen. However, unlike the choice-focused adult worker model, this model stresses that women's participation in the labour market is a private matter much more than it is a public concern. Hence, the government in this adult worker model should avoid direct support for women to take part in the labour market through pro-employment measures, instead focusing on assisting those women who are unwilling or unable to take part in the labour market by providing a reasonable standard of living through decommodification measures.Footnote ²

The desirability of the four types of the adult worker model varies in terms of their capacity to reduce gender inequality and enhancing women's freedom to choose the ways of participating in the family. Given that the market-focused model is associated with government minimum intervention, its effectiveness in reducing the gender inequality in the allocation of paid work in the work economy and that in the allocation of unpaid work in the family should not be over-estimated. Moreover, women are unlikely to receive sufficient support from the government (for example, through the provision of family–work reconciliation measures) to explore more choices concerning participation (or non-participation) in the family. With the emphasis on promoting women's employment, the supported model is unlikely to give women sufficient support to choose to be a full-time family carer. Unlike the supported model, the choice-focused model stresses the importance of enhancing women's freedom to choose between taking part in the work economy as a worker and performing the role of full-time family carer. If the government tries to strengthen this model through encouraging more men to take up the unpaid work in the family, it also reduces the gender inequality. As the collective consumption model emphasises supporting women to lead a decommodified life, it does not attach importance to the provision of support to women for developing their career or make important impact on reducing the gender inequality in the allocation of paid work and unpaid work.

Hong Kong

This section is concerned with the application of the adult worker models in the analysis of the policy responses to the defamilisation and familisation risks in the empirical world. Our focus is on the gap between the pure form of the adult worker model in the literature and the actual form of the adult worker models upheld by the government in practice. In order to illustrate our views on these issues, we discuss the major pro-employment and decommodification measures provided in Hong Kong.

There are reasons for selecting Hong Kong as a case example for discussion. Evidence shows that the Hong Kong government may not attach its policy to exclusively a single type of adult worker model. Successive administrations in Hong Kong have stressed low direct tax policy and their commitment to containing the growth of public expenditure as a proportion of GDP, meaning it has long been regarded by international organisations as a champion of economic freedom (Miller and Kim, Reference Miller and Kim2017). It is not unusual for the Hong Kong government to stress the importance of avoiding too much intervention in society lest a dependency culture would be encouraged, market forces would be disrupted and the fiscal sustainability of the Hong Kong political economy would be jeopardised (Social Welfare Advisory Committee, 2010). This evidence suggests that the Hong Kong government is likely to support the market-focused adult worker model.

In 2014, the then Chief Executive C.Y. Leung introduced a new ‘poverty alleviation policy […] to encourage young people and adults to become self-reliant through employment’ (Reference LeungLeung, 2014). The so-called Low-income Working Family Allowance suggested increased welfare payments only for those claimants that also work longer hours. In the same year, the Hong Kong government has consulted the public concerning the ways of increasing women's labour participation rate in 2014 (HKSAR Government, 2014) and most recently, the new Chief Executive, Carrie Lam, in her Reference Lam2017Policy Address: We Connect for Hope and Happiness, pledged further efforts by the Hong Kong government ‘to protect the interests and well-being of [Hong Kong] women and unleash their working potential’ (Reference LamLam, 2017: 53). These policy ideas seem to favour more the supported adult worker model than the market-focused adult worker model.

The mixed messages made by the Hong Kong government suggest that the adult worker model supported by its policies may be more complicated than can be reflected in one of the above four adult worker models. For this reason, it is believed that Hong Kong provides a favourable ground for examining the gap between the ideal and actual form of the adult worker models.

Conclusion

With the focus on the connection between the studies of defamilisation/familisation and the adult worker models, this paper has presented three analytical tasks. Firstly, four adult worker models have been suggested based on different combinations of the pro-employment and decommodification measures. Secondly, we have discussed the relative desirability of these four models in tackling the defamilisation and familisation risks faced by women, and in turn enhancing women's well-being. Thirdly, with the focus on the case example of Hong Kong, we have pointed out the gap between the pure form of the adult worker models, and the actual form of the adult worker model upheld by the government. The relationship between the decommodification and pro-employment measures, the adult worker models and the defamilisation and familisation risks faced by women is shown in Figure 1.

Figure 1. A chain of relationships.

As shown above, the choice-focused adult model is more desirable than the other three models in terms of promoting gender inequality and giving women's freedom to choose the way of participating (or not participating) in the family. However, this does not necessarily mean that it would be favoured by the government. No evidence suggests that the Hong Kong government is keen to put this model into practice. Instead it develops its own model with reference to the supported and market-focused adult worker models. Given the inadequacies of the supported and market-focused adult worker models in promoting women's freedom to choose the roles to perform inside and outside the family and enhancing gender equality, the keenness of the Hong Kong government to safeguard women's well-being should not be over-estimated.

The discussion of the case example of Hong Kong shows the value of discussing the four adult worker models together in analysing the policy responses to women's defamilisation and familisation risks. With reference to these models, we can identify not only the policy instruments used by the government to assist women to respond to the demands of the paid labour market but also the effects of the implementation of these instruments on women's well-being (in terms of the gender equality and women's freedom to choose the ways of participating in the family). Moreover, we can suggest a number of policy alternatives based on the four adult worker models or different combinations of these models.

In concluding this paper, we may take the opportunity to suggest a possible agenda for further research. This paper has discussed four pure forms of the adult worker model. It is worth doing more investigation into the applicability of these pure forms into the empirical world. As shown above, the discussion of Hong Kong shows that the relevance of the market-focused and supported models to those governments which favour residualism but also want to take a more active role in encouraging women to take part in the work economy. More work could be done to find out whether we have governments which are keen to uphold the other two models. Since the scope of this investigation is large, it may not be covered by a single paper.

Moreover, in addition to the search for ways to support women to choose different roles to deliver both inside and outside the family, it is worth exploring the ways for promoting ‘men's choices’. As shown above, there is a possibility that promoting women's choices can reinforce men's choices – for example, through giving men more time rights to do the unpaid work in the family. However, there is also a possibility that promoting women's choices can limit men's choices (Giullari and Lewis, Reference Giullari and Lewis2006) – for example, through legally obliging men to do more unpaid work in the family. Hence, for enhancing both men's and women's well-being, it is important to conduct several analytical tasks, including exploring an ideal combination of the decommodification and pro-employment measures that can assist both women and men to reduce their defamilisation/familisation risks, examining whether and how this combination of measures can enhance both men's and women's choices, and investigating why governments favour this combination of the pro-employment and decommodification measures or why they do not. By carrying out these analytical tasks, we can include men's life in the feminist debates as suggested by (Sung and Pascall, Reference Sung, Pascall, Sung and Pascall2014). Hence, it is worth addressing these analytical tasks in future research endeavours.