Female Labour Market Participation, Social Policy and the Gender Care Gap. A Fuzzy-Set Analysis
Barbara Da Roit*, Marcel Hoogenboom*and Bernhard Weicht*
* Faculty of Social Sciences, UtrechtUniversity, the Netherlands
Correspondence address: .
FIRST DRAF: PLEASE DO NOT QUOTE WITHOUT PERMISSION OF THE AUTHORS
Abstract
In this paper we investigated the relationship between the relative contributions of women – in comparison to men: the “gender care gap” – to the informal care for their partners, government care policies and the labour market position of European women. We analysed the interlinkages between the three factors in thirteen European countries by applying a research model based on the Fuzzy set/Qualitative comparative analysis method. We chose this approach in order to be able identify patterns in the relationship between the three factors, since in the literature it is suggested that none of the factors alone can explain the presence or absence of the gender care gap. The results of our analysis suggests that the extent to which countries are confronted with the ageing of their populations and the (absence of a) policy reactions to this problem are crucial to understanding the gender care gap in the care for elderly parents in need of care, while the presence or absence of gendered care attitudes and labour market characteristic play a relatively minor role.
Keywords:
Elderly care; gender care gap; labour market; social policy; fuzzy-set analysis
1. Introduction
The gender gap in family caregiving is an established research finding: men tend to dedicate less time and energy than women to caregiving and to provide specific gendered types of help. Yet while the genderedness of caregiving has been well documented in the literature, its explanations are less conclusive. In the literature roughly three alternative explanations can be distinguished. One explanation seeks the roots of the “gender care gap” between men and women in (the interiorization of) gender norms, while a second one explains the gap by pointing to differences between the positions of men and women on the labour market. Finally, a third type of explanation relates the gender care gap to (the absence of) welfare state policies. Following recent claims in literature that the gender care gap should be explained by a combination of the above mentioned factors, in this paper weuse Fuzzy set/Qualitative comparative analysis (Fs/QCA) (Ragin 2000, Kvist 2007) to identify patterns in the relationship between gendered care giving for elderly people (75+), labour market position of men and women and the availability of government care arrangements for elderly people in thirteen European countries. After discussing the theoretical background (Section 2) and specifying the methodology of the research (Section 3) we present our findings (Section 4). Finally, in Section 5 we highlight the contribution of the research to the debate, its limitations and suggestions for future investigation.
2. Theoretical background
2.1 Gendered division of care
Gendered patterns of care-giving have been of interest for a long time (Neal et al. 1997). However, these practices have largely been described as representing gender patterns and cultures and cross-national variations in the gender division of care are often explained through a focus on normative dimensions of welfare policies which are themselves mainly based on societal gender cultures (Aboim 2010).It has been established in the literature that in the context of elder care women fulfil the larger part of care-giving in most countries. In a cross-national study Bracke et al. (2008) show that the spouse is the number one caregiver in all age categories and that especially women fulfil care activities when their partner is ill. While the share of men caring for an elderly person increases dramatically with age (Kahn et al. 2011, Del Bono et al. 2009, Cahill 2000) men’s caring involvement is predominantly linked to spouses (Bracke et al. 2008). Dahlberg et al. (2007) also confirm that the time spent caring increases with age, with the highest levels of caring commitment in people aged over 80 years, and that overall women commit more time to caring than men. The authors also indicate, however, that men take over a large proportion in later life (caring for their partners), a time when people have usually left the labour market. Also in the type of care delivered a gender division can be found. Many studies report that women fulfil more intense, personal care and often provide also more managerial care (Rosenthal et al. 2007, Kruijswijk, Da Roit and Hoogenboom, forthcoming) both in countries with a more equal gender division, such as Sweden (Jegermalm 2006) and countries with a stronger divide, such as Japan (Hanaoka and Norton 2008) or Italy (Toffanin 2011). In terms of consequences of caring Rosenthal et al. (2007) report that the provision of care often leads to increased stress for women while for men care delivery often interferes with work commitments.
Apart from different gender cultures how can the discrepancy between men and women’s caring practices be explained? Geist (2005) argues that while occupational gender segregation, gender differences in wages and workplace authority and the gender gap in poverty have received widely attention, the factors determining stratification processes within families have been studied less and that, in particular, an integration of both perspectives within one framework is missing.
2.2 Structural factors
It has widely been argued that the gendered division of care labour cannot sufficiently be explained by individual decisions but are rather shaped by structural factors linked to economic, political and social conditions (Cooke 2006). Several structural factors have been identified in shaping the gendered division of household labour and care labour. The access to and the use of long-term care services varies according to health status, gender and socio-economic position (Portrait et al. 2000). Similarly, in terms of support for informal carers, Rogero-Garcia and Rosenberg (2011) observe that support (paid and unpaid) is significantly lower among households with female caregivers and that in particular the lower educated caregivers receive little outside support. Also Ungerson (2000) emphasises the importance of economic and structural factors in determining the distribution of care labour. In particular she argues that gender on its own is not sufficient as a social category to understand the social division of care. Rather, economic inequalities linked to jobs and wages are equally important in creating the link between labour market inequalities and care inequalities, both between men and women, and between women from different social economic backgrounds.
Arksey and Morée (2008) identify an important tension between the fact that several countries introduce policies to increase the participation of disadvantaged groups on the labour market (such as women and older workers) while at the same time demanding more from people in terms of care giving. This tension between the demands of employment and the requirement of caring (see Crompton 2001) carries particular gendered connotations, in that gender equality is promoted in the context of the labour market, while the gendered division of care work often remains unchallenged (MacLeavy 2007). Öun (2012) and Meiko (2010) therefore emphasises the important role welfare institutions can also play in the household division of labour.
Thus, two structural factors specifically influence the realm of private care giving and the division of care labour within people’s households: the labour market conditions and structures and the social policies in place.
2.2.1 The labour market
In the literature so far inequality in the division of housework has often been linked to labour market inequality where labour market differences are usually seen as consequences of domestic labour (Geist 2005). This causal relationship is assumed in much of the literature studying the gendered division of care labour. Jang et al. (2012), for example, using SHARE data, emphasise that intensive care negatively affects labour force participation among middle aged women in both northern and southern European countries. They conclude that introducing further formal arrangements for care which allow in particular women to share their informal carer’s roles might lead to an increase in the participation of women in the labour market. Knijn and Kremer (1997) establish a framework to analyse the relationship between labour market and care obligations, inherently linked to gendered discussions of care as public or private possibility and paid or unpaid work. They particularly focus on the development of social policies, originally designed to ‘liberate people from the obligation to work so that they could care’ (329). In the context of child care the relationship between the participation in the labour market and care responsibilities has been studied widely. Ellingsæter and Gulbrandsen (2007) argue that, in the case of Sweden or Denmark, the development of well-funded public child care was explicitly linked to the facilitation of mothers’ employment and gender equality. However, political discourses and policies still remain to imagine women to be the main provider of informal care (Outshoorn 2002).
Spiess and Schneider (2003) identify a negative relationship for women between the changes in care giving and changes in working times in European Union countries arguing that a change in work hours is negatively associated with the beginning of a care giving period. In their analysis they could not, however, establish a relationship between the ending of a care giving period, or the reduction of care hours and increasing labour market participation. The authors argue that this suggests that reductions or terminations of work due to care giving responsibilities are not compensated later. They also find important differences between northern European countries and Southern European countries (including Ireland), linking the consequences of care giving on working hours to the availability of formal care arrangements.
Similarly, Kotsadam (2011), drawing on data from the ECHP, finds women’s employment to be negatively associated with care-giving to elderly people. However, the gendered effects seem to be more negative in Southern European countries and less negative in the Nordic countries, and in between in the Central European countries. He demonstrates that not only do women provide more care in Southern European countries but that also the negative correlation with the probability of being employed and the number of working hours is stronger. Other authors, however, argue that a straightforward link between women’s increased labour force participation and reduction of informal care cannot be observed (Doty et al. 1998), as care responsibilities are often taken over at times when people have already left the labour force.
In an analysis of the weaker labour market involvement of informal carers in the UK Henz (2006) concludes that the lower levels of involvement of female carers to the labour market was predominantly a result of their adaptation to caring. This means that part-time working women are more likely to leave the labour market than those working full-time. The reason she finds not in the idea that part-time work is easier to combine with care obligations but rather that part-time work is seen as marginal work and temporary work in comparison to full-time work. However, also this link is influenced by socio-economic differences in that rather working class women reported effects of caring on their work arrangements. In another contribution Henz (2010) applies a household-division-of-labour perspective to analyse the distribution of care, showing that those working full-time provide lower shares of care for their parents with needs than those who are less strongly attached to the labour market. Henz (2010) argues that this effect is not only linked to time constraints of carers but rather higher bargaining power derived from paid works.
Elsewhere, Henz (2009) argues that in particular for men full-time employment seems to be a reason not to engage in care. However, also the distribution of fulltime and part-time work is only one factor in the link between labour market characteristics and care divisions. Tavora (2012) shows with the case of Portugal that high rates of female fulltime employment can also be linked to high levels of familialism in the realm of care. Putten et al. (2010) in a study on the Netherlands find no empirical relationship between work hours and the provision of support to parents. They therefore argue that the feared declining support for older people as a result of changing labour market participation for women cannot be observed in the Netherlands. For men they find a particularly inelastic relationship between work hours and caring responsibilities (men do not cut down working hours as a result of supporting parents in need). However, the Dutch case is particular in the sense that part-time work is widely available and that in middle age the vast majority of women (and increasingly men) work part-time, which might enable an easier combination for both men and women of working and caring responsibilities.
Sarkisian and Gersel (2004) specifically analyse the effects of different employment characteristics (and differences between men and women) on the gender gap in helping parents. They find that apart from the crucial factor of having or not having a job (which in particular counts for women’s availability to care) certain job characteristics can explain the gender gap. While the analysis suggests that, all things being equal, employed women and men provide equal amounts of help the authors strongly emphasise the fact that employment patterns differ significantly and that therefore, all things are not equal. In particular they identify different wage levels and the number of self-employed people as defining the gender gap. As higher wages are typically associated with fewer hours of supporting parents the gender wage gap translates itself into a gender care gap. Similarly, the fact that fewer women are self-employed leads to a higher probability to take over the help for parents. Henz (2010) agrees that spouses do not heavily negotiate the care for their parents but that rather employment conditions play a crucial role. In particular full-time employment, wage differences (in particular in low-income households) drive the gender care division (see also Heitmueller and Inglis 2007). While Sarkisian and Gersel (2004) therefore recommend that a change in the payment structures of jobs would directly change gender divisions in people’s support of their parents, they do not take into account the role social policies can play in this equation.
2.2.2 Social policies
Social policies can be found in relation to both, labour market and care. Generally social policies for supporting unpaid care work have developed rather modestly all over Europe when compared with labour market activation policies (Pascall and Lewis 2004). Bühlmann et al. (2010) argue that a comparative perspective shows that the division of labour is different in countries with more egalitarian values and that this shift in values is strongly moderated by welfare policies. These policies are furthermore linked to equality policies on a broader scale. Fuwa and Cohen (2007) in that context observe how women’s fulltime employment and higher income has strong effects on the gender division of housework in countries with greater equality of access policies.
At the same time it is important to pay attention to the fact that policies are also based on an image of and that they also reproduce images of women and men in certain social roles (Rake 2001). Hammer and Oesterle (2003) for example, show the reproductive effect of (social) policies on the gendered division of care labour arguing that in Austria only for those with higher financial means alternatives of informal care are accessible, and Morgan and Zippel (2003), demonstrate how paid child care leave policies are reinforcing the traditional division of care work in people’s homes, also observed by Kvande and Brandth (2009) in relation to cash for care systems.
The fact that the design of the welfare state and its (care) policies is affecting both women’s labour force participation (Semyonov 2006) and the gendered division of informal care labour (Ungerson 2000) requires a framework that links together the effects of public interventions and policies on both fields of interest (Saraceno and Keck 2011). Therefore the question we aim to answer in this paper is: is there a relationship between government care policies and the labour market position of European women on the one hand, and the relative contributions of women (in comparison to men: the “gender care gap”) to the informal care for partners on the other? And if so, what patterns between countries can be identified how can these patterns be explained? In order to be able to investigate the interlinkages between the three factors – gender care gap, labour market position and social policies – in the next section we will develop a research model based on the Fuzzy set/Qualitative comparative analysis methodwhich enables the identification of patterns in the relationship between the three factors in thirteen European countries.