ABSTRACT
In this study, we investigate couples' division of household tasks by the age of the youngest child, comparing France, eastern Germany, and western Germany. For our analyses, we draw on Generations and Gender Survey (GGS) data. As expected, our findings are that the division of housework is less egalitarian for couples with preschool age children than for childless couples, and these differences are greatest in western Germany. However, we had also expected the division of housework to be more egalitarian again for couples with older children, among whom maternal employment rates are higher than among those with younger children. Our findings confirmed this expectation for western Germany. Surprisingly though, we found that in both eastern Germany and France, the division of housework was actually continuously less egalitarian the older couples' children. An explanation may be that the traditionalizing impact of parenthood unfolds slowly with parenthood duration as couples increasingly yield to societal expectations regarding parental roles. In western Germany, where women reduce their employment most significantly when becoming mothers, employment status effects appear to dominate any other trends associated with the age of the youngest child.
1. Introduction
In this study, we compare couples' division of household tasks in France, western Germany, and eastern Germany. While previous studies have conducted general international comparisons (Davis and Greenstein 2004; Fuwa 2004; Geist 2005), we focus specifically on cross-national differences in housework allocation by age of the youngest child.
Comparing parents' division of housework between France and Germany should be particularly interesting. France and Germany have both been described as corporatist, continental European welfare states (Esping-Andersen 1990). Yet, the gender regimes in both countries differ in essential ways, with France characterized by Lewis (1992) as a modified male breadwinner country, and Germany fitting the description of a strong male breadwinner country.
Moreover, Germany cannot be studied as a single unit of analysis. In the post-war period, East and West Germany adopted contrasting sets of family policies. Those in East Germany encouraged high female employment rates, while West German policies strongly supported the male breadwinner model of the family. Even today, the childcare infrastructure in eastern Germany is far more extensive and more closely resembles that in France than that of western Germany (Evers et al. 2005; Lewis et al. 2008). These differences are crucial for our comparative study, as childcare especially influences mothers' employment, which in turn can be expected to influence couples' division of housework (South and Spitze 1994).
Our findings show that the low employment rates among mothers of young children in western Germany are mirrored in a very unequal division of housework. In eastern Germany and France, women do not reduce their employment as significantly upon entry into parenthood as in western Germany, and the division of housework among parents of young children is also more egalitarian. However, while we find the expected return to a more egalitarian division of housework for couples with older children in western Germany, among whom maternal employment rates rise again, the division of housework actually becomes less egalitarian for those with older children in France and eastern Germany. Presumably, couples initially make an effort to maintain an egalitarian division of housework on becoming parents, but then increasingly yield to societal expectations regarding more traditional parental roles as children grow older.
The structure of the paper is the following. The next section discusses differences in cultural models of the family as well as in the institutional setting in France, eastern and western Germany (section 2). We then turn to theoretical considerations on couples' division of housework (section 3). After presenting our research questions (section 4), we discuss the data and method in section 5. In section 6 we then present the results of our empirical analyses. The paper closes with a summary and concluding discussion of our main findings.
2. Institutional background
France, eastern and western Germany differ strongly in terms of the culturally predominant family model that has influenced the development of family policies, and in the historical pathways that led to an orientation towards each respective type of family model.
According to Pfau-Effinger (2004), the specific pattern of quick industrialization, urbanization, and large-scale rural to urban migration in Germany at the turn of the nineteenth to the twentieth century increased the importance of the urban bourgeoisie as a role model. This led to widespread societal acceptance of the housewife/male breadwinner model of the family endorsed by this social class. In the post-war period, welfare state institutions in West Germany further supported this culturally widely accepted family model (Pfau-Effinger 2004). These include tax benefits in favor of married couples where one partner is not working (Schratzenstaller 2002). Health insurance is also free of additional costs for non-working spouses of the employed. Moreover, there has historically been very little state support for childcare outside the home (Evers et al. 2005). By the 1980s and 1990s, however, attitude surveys showed that the male breadwinner/female part-time carer model had become the most popular family model. Most likely, contradictions between ideals of equal citizenship in a modern society and the housewife model of the family had become too great for the housewife model to remain culturally desirable (Pfau-Effinger 2005). Very recently, parental leave and childcare policies giving more support to dual earner families have been passed in Germany (Lewis et al. 2008). However, these changes are too recent to impact the findings in this study.
In the post-war period, family policies in East Germany by contrast strongly diverged from the historical male breadwinner tradition. The institutional framework was instead set to encourage a dual earner model of the family. This was motivated partially by goals of enhancing economic development, and partially by socialist ideals (Pascall and Kwak 2005). The childcare infrastructure was very extensive, parental leave policies served to improve work–family compatibility, and wage levels were generally too low for families to live on only one income (Rosenfeld et al. 2004; Pfau-Effinger and Smidt 2011). Yet, the measures for work–family compatibility passed by the East German regime did not emanate from broadly shared goals at the societal level, and the division of housework remained largely uncontested (Pascall and Lewis 2004). Nonetheless, younger generations have come to value possibilities for mothers to be employed full-time, and to make use of externally provided childcare, even after the institutional changes brought about by unification (Pfau-Effinger and Smidt 2011). While the former West German tax and transfer systems universally apply after unification, important differences in the childcare infrastructure remain between eastern and western Germany. Childcare is organized at the municipal level (Evers et al. 2005), and although there were cutbacks in childcare provision in eastern Germany, municipalities still opted to maintain substantially higher levels of childcare provision than in western Germany. In eastern Germany, 38% of all under-3-year olds attended public day care in 2008, compared to 10% in western Germany. Among children aged 3–6, 62% attended full-time childcare in eastern Germany, while only 20% did so in western Germany (Statistische Ämter des Bundes und der Länder 2009). This is also reflected in regional differences in female employment rates (Forsberg et al. 2000).
Pfau-Effinger (2005) describes France as belonging to a group of countries in which the dual breadwinner/external childcare model is the culturally dominant family model, and Lewis (1992) describes France as a modified, rather than a strong, male breadwinner country. Historically, however, ideals of gender equality do not seem to have been the most influential factor for developments in family policy. Instead, there has been a focus on ameliorating inequalities between the childless and families with children, which were partially motivated by pronatalist concerns (Lewis 1992). The tax system benefits households with children, and additional benefits are provided independent of income for families with more than one child (Fagnani 2006). There has historically been little opposition to maternal employment, and maternity leave was introduced as early as 1913 (Lewis 1992). France has had a long history of external childcare provision. Free nursery education is available for all children aged 3–6 (Windebank 2001), and take-up is practically universal (OECD 2010). Crèche spaces were expanded significantly in the 1980s, partially motivated by the expansion of the tertiary sector (Martin et al. 1998; Lewis et al. 2008), and in 2006, 43% of children aged under three attended formal care (OECD 2010).
Implications of these international and regional differences in family policies for maternal employment patterns are well documented (see for instance Kreyenfeld et al. 2009). We give an impression of employment levels for women with a partner by age of the youngest child (Table 1), as our main analyses in section 6 will focus on couples. As can be seen here, the difference in the percentage working full-time between childless women and mothers of a child aged 0–5 is smallest in France and greatest in western Germany. In western Germany, the percentage of mothers working full-time is consistently lower than in eastern Germany or France at all ages of the youngest child. In western Germany, it appears to be especially common to work short part-time hours of less than 20 hours a week.
. | Employment status . | ||||
---|---|---|---|---|---|
Age of the youngest child . | >=35 hours . | 21–34 hours . | <=20 hours . | Not employed . | Student . |
France | |||||
No children | 59% | 9% | 4% | 16% | 13% |
Age 0–5 | 45% | 16% | 9% | 29% | 1% |
Age 6–13 | 46% | 19% | 10% | 26% | 0% |
Age 14–17 | 46% | 14% | 13% | 27% | 1% |
Age 18+ | 50% | 13% | 8% | 29% | 0% |
Non-resident children only | 44% | 13% | 10% | 33% | 0% |
Eastern Germany | |||||
No children | 42% | 14% | 2% | 27% | 14% |
Age 0–5 | 24% | 11% | 7% | 56% | 2% |
Age 6–13 | 39% | 28% | 4% | 30% | 0% |
Age 14–17 | 52% | 24% | 0% | 24% | 0% |
Age 18+ | 53% | 18% | 2% | 25% | 1% |
Non-resident children only | 41% | 20% | 6% | 33% | 0% |
Western Germany | |||||
No children | 52% | 11% | 9% | 17% | 11% |
Age 0–5 | 13% | 15% | 10% | 60% | 1% |
Age 6–13 | 17% | 25% | 22% | 35% | 0% |
Age 14–17 | 28% | 26% | 22% | 24% | 1% |
Age 18+ | 32% | 26% | 14% | 28% | 0% |
Non-resident children only | 34% | 18% | 15% | 33% | 0% |
. | Employment status . | ||||
---|---|---|---|---|---|
Age of the youngest child . | >=35 hours . | 21–34 hours . | <=20 hours . | Not employed . | Student . |
France | |||||
No children | 59% | 9% | 4% | 16% | 13% |
Age 0–5 | 45% | 16% | 9% | 29% | 1% |
Age 6–13 | 46% | 19% | 10% | 26% | 0% |
Age 14–17 | 46% | 14% | 13% | 27% | 1% |
Age 18+ | 50% | 13% | 8% | 29% | 0% |
Non-resident children only | 44% | 13% | 10% | 33% | 0% |
Eastern Germany | |||||
No children | 42% | 14% | 2% | 27% | 14% |
Age 0–5 | 24% | 11% | 7% | 56% | 2% |
Age 6–13 | 39% | 28% | 4% | 30% | 0% |
Age 14–17 | 52% | 24% | 0% | 24% | 0% |
Age 18+ | 53% | 18% | 2% | 25% | 1% |
Non-resident children only | 41% | 20% | 6% | 33% | 0% |
Western Germany | |||||
No children | 52% | 11% | 9% | 17% | 11% |
Age 0–5 | 13% | 15% | 10% | 60% | 1% |
Age 6–13 | 17% | 25% | 22% | 35% | 0% |
Age 14–17 | 28% | 26% | 22% | 24% | 1% |
Age 18+ | 32% | 26% | 14% | 28% | 0% |
Non-resident children only | 34% | 18% | 15% | 33% | 0% |
Weighted, N= 4834.
Source: GGS 2005, own calculations.
Differences in employment patterns are reflected in differences in couples' relative earnings as well. For instance, Trappe and Sørensen (2006) show that women's earnings constitute a greater proportion of household income in eastern than in western Germany. Extended periods of non-employment or part-time employment can lead to lower hourly wages as well (Ziefle 2004; Fouarge and Muffels 2009). Most studies find a greater motherhood hourly wage penalty in Germany than in France or find that a larger raw gender wage gap in Germany than in France can be explained by taking various factors such as employment experience into account1 (Davies and Pierre 2005; Arulampalam et al. 2007).
3. Theoretical considerations
Normative expectations and attitudes on gender roles acquired through socialization are likely to be important factors influencing couples' decisions on the division of housework (Hiller 1984; Windebank 2001). The different cultural models of the family described in the previous section do indeed appear to be reflected in people's attitudes. Analyzing Eurobarometer data for 2006, Scheuer and Dittmann (2007) for instance show that 20% of eastern German, 32% of French, and 53% of western German respondents agreed with the statement that ideally, women should stay at home and take care of the children while men should go to work. Attitudes on parental roles do however appear to have become more liberal across time. For instance, in 1991, 76% of western and 58% of eastern Germans still agreed with the statement that young children will suffer if their mother is employed, compared to only 60 and 23% in 2006, respectively.
Increasingly liberal attitudes towards gender roles may be one explanation for Kaufmann's (1994) finding that nowadays, many young couples no longer have clear expectations regarding the division of housework at the beginning of their partnership. When children are born, the significant increase in the amount of tasks to be accomplished however makes it particularly advantageous to establish routines (Kaufmann 1999). For these routines, parents often revert to socially sanctioned gender roles. Kaufmann (1994) also holds that, due to differences in socialization, women frequently possess more knowledge regarding many household tasks and do not always trust their partners to carry them out well enough to serve the baby's needs.
Other, predominantly rational choice-based theoretical approaches indicate that in contexts where employment and childcare is not well compatible, changes in mothers' employment status around the birth of a child can be expected to lead to housework reallocations. The time availability approach, for instance, predicts that the partner who spends less time in paid employment does a larger share of the housework (South and Spitze 1994; Bianchi et al. 2000). The relative resources approach, however, maintains that couples allocate more housework to the partner with lower economic resources in terms of income or education, either because this represents a more efficient division of labor, or because the partner with lower resources is in a weaker bargaining position (Becker 1981; Hiller 1984).
In any case, previous empirical studies have indeed found entries into parenthood to be related to changes in the direction of less egalitarian divisions of housework (Sanchez and Thomson 1997; van der Lippe et al. 2010). We will study how this parenthood effect is related to the age of the youngest child and how this differs internationally. We would expect that parenthood causes greater changes in the division of housework in western Germany than in France or in eastern Germany, since changes in women's employment status, as well as adherence to traditional attitudes regarding mothers' roles, are strongest in western Germany.
Further factors which, on theoretical grounds, can be expected to be linked to the division of housework are included in our analyses as well. For instance, cohabitation implies less security in the case of separation, and cohabiting partners may therefore be less inclined to specialize in specific tasks than are married couples (Batalova and Cohen 2002). Many couples marry around the time of entry into parenthood (Perelli-Harris et al. 2012). Thus, part of the parenthood effect may be related to differences in marital status.
Not only marital status, but stepfamily status, too, may be related to couples' division of labor. It has been argued that for second marriages, fewer normative expectations exist with regard to family life (Ishii-Kunz and Coltrane 1992; Baxter 2005) and negative experiences with the division of housework in first marriages may induce people to rethink this aspect of family life in their next marriages (Sullivan 1997). Many, though not all, stepparents are in a second partnership, and thus these arguments may hold for stepfamilies as well. Previous research has indeed found that housework is divided more equally in stepfamilies than in families with common biological children only (Snoeckx et al. 2008). At the same time, it seems plausible to assume that the division of childcare tasks in stepfamilies may be very different from the division of housework, with stepparents contributing less to childcare than biological parents. Since the focus of our study is on housework though, we do not look into these differences.
A further factor that can be expected to be related to couples' division of housework is partnership duration. The division of housework has been hypothesized to become more unequal the longer the duration of the partnership (Schulz and Blossfeld 2006; Wengler et al. 2009). While partners may at first put effort into dividing housework equally, they increasingly yield to traditional gender role expectations at longer partnership durations.
4. Research questions
We aim to investigate whether differences in the division of household labor between parents and childless couples are greater in western Germany than in France or eastern Germany. As discussed earlier, employment rates among mothers of preschool children are lower in western Germany than in eastern Germany or France. Thus, differences in the division of household labor between parents and childless couples should also be greater in western Germany than in eastern Germany or France, according to the relative resources and time availability approaches. Due to the male breadwinner tradition in western Germany, stronger normative expectations linking motherhood to domesticity might also contribute to a greater motherhood effect on the division of housework there.
Next, we are interested in differences in the division of household labor between parents of children of different ages in each of the three countries or regions. We expect that in France and eastern Germany, the division of housework is nearly as egalitarian even for couples with young children as for childless couples. This is because mothers' employment breaks are fairly short. For western Germany, we would expect a much less egalitarian division of housework for couples with younger children and a return to a more egalitarian arrangement only for those whose children are considerably older.
5. Data and method
For our analyses, we draw on Generations and Gender Survey (GGS) data. The internationally comparable Generations and Gender Surveys are coordinated by the Population Activities Unit (PAU) of the United Nations Economic Commission for Europe (UNECE).2 The Gender and Generations Surveys focus on the fields of fertility and family dynamics, home-leaving, and retirement (United Nations 2007). The GGS is designed as a panel survey. So far, data from wave 1 is available for analysis for both Germany and France. Interviews for wave 1 were conducted in 2005 in France and Germany. Our analyses thus refer to the year 2005.
For our dependent variable, we made use of respondents' replies to items indicating whether they themselves or their partner are mainly responsible for four different everyday household tasks. These include cooking, doing the dishes, vacuum-cleaning, and grocery shopping. We combined responses to the items on the different household tasks following the method suggested by Wengler et al. (2009). We assigned a score of 1 if the respondent indicates that the man is mainly responsible for a given household task, a score of 0 if both share the task equally, and a score of −1 if the woman is mainly responsible for this task. We then summed up these scores for each respondent, and divided by 4 so that our indicator varies between −1 and 1. We apply linear regression analysis to investigate determinants of the division of everyday household tasks.
A problem with the data is of course that it is cross-sectional. We cannot be completely certain that differences in the division of housework between childless couples and parents indicate a causal effect of parenthood. Couples with a more traditional division of housework might self-select into parenthood. Panel data on couples' division of housework is rarely available, particularly for internationally comparative studies. Thus, most studies on couples' division of housework are based on cross-sectional data (e.g. Batalova and Cohen 2002; Davis and Greenstein 2004; Fuwa 2004; Baxter 2005; Geist 2005). The few single-country analyses that did make use of panel data have however shown that entries into parenthood do strongly affect the division of housework (Sanchez and Thomson 1997; Schulz and Blossfeld 2006). Moreover, Sanchez and Thomson (1997) show that among childless couples in the first wave of their study, there are almost no differences in the division of housework between those who will become parents and those who will not become parents by the next wave. The transition to parenthood then however strongly affects the division of housework.
The main focus of our research is not on entries into parenthood, but on the effect of the age of the youngest child on parents' division of housework. Our main analyses do include childless couples as a basis of comparison. However, we have conducted additional analyses restricting our sample to parents, and we find very similar results. We ran further robustness checks in which we limited our sample to a narrow age range of couples where the woman was aged 34–49. This was the 15-year age range that maximized the included number of parents in each of the groups with children aged 0–5, 6–13, 14–17, and over 18. Our results remained substantively the same for this restricted cohort, indicating that they are not predominantly driven by generational effects. We control for age in all models. In the model for the restricted age range, no significant effect of age was found.3
6. Results
Our descriptive results, shown in Figure 1, indicate that women do the greater part of everyday household tasks in all three countries or regions. A value of 0 indicates that couples share everyday household tasks equally, values between 0 and −1 indicate that women do a larger share, and values between 0 and +1 indicate that men do a larger share. As can be seen here, the division of housework among childless couples in western Germany is slightly less egalitarian than is the case in eastern Germany or France. Possibly, more traditional gender role expectations lead to a less egalitarian division of housework even among the childless, where there are not yet any pronounced international differences in employment rates. Among parents, women's share of everyday household tasks is generally greater than for childless couples and this difference is largest in western Germany.
Division of everyday household tasks among childless couples and parents Descriptive results
Source: GGS 2005, own calculations.
Source: GGS 2005, own calculations.
Figure 2 and Table 2 give results for the division of everyday household tasks by age of the youngest child from our first model.4 As was the case for the descriptive results, a value of 0 would indicate that couples share everyday household tasks equally, values between 0 and −1 indicate that women do a larger share of these tasks, and values between 0 and +1 indicate that men do a larger share of these tasks. We had expected the division of household tasks to be less egalitarian for parents of young children than for childless couples, and for this difference to be greatest in western Germany. This is indeed what we find. Among parents of children aged 0–5, women do a greater share of everyday household tasks than among childless couples, and this difference is greatest in western Germany. This is likely to be related to differences in employment patterns as well as role expectations for mothers in western Germany as compared to eastern Germany and France (Table 1).
Division of everyday household tasks by age of the youngest child
Results from model 1
Absolute values for division of everyday household tasks (at the reference categories of the control variables).
Source: GGS 2005, own calculations.
Results from model 1
Absolute values for division of everyday household tasks (at the reference categories of the control variables).
Source: GGS 2005, own calculations.
. | France . | Eastern Germany . | Western Germany . | ||||||
---|---|---|---|---|---|---|---|---|---|
. | Model 1 . | Model 2 . | Model 3 . | Model 1 . | Model 2 . | Model 3 . | Model 1 . | Model 2 . | Model 3 . |
Age of the youngest child | |||||||||
No children | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Age 0–5 | −0.11*** | −0.05** | −0.05* | −0.13* | −0.05 | −0.04 | −0.21*** | −0.08** | −0.05 |
Age 6–13 | −0.16*** | −0.12*** | −0.11*** | −0.10 | −0.05 | −0.04 | −0.24*** | −0.13*** | −0.11*** |
Age 14–17 | −0.21*** | −0.17*** | −0.14*** | −0.16** | −0.12* | −0.11 | −0.15*** | −0.05 | −0.03 |
Age 18+ | −0.27*** | −0.22*** | −0.19*** | −0.29*** | −0.24*** | −0.22*** | −0.09*** | −0.03 | −0.02 |
Non-resident children only | −0.19*** | −0.15*** | −0.13*** | −0.06 | −0.02 | −0.01 | −0.05* | 0.00 | 0.00 |
Number of children | |||||||||
1 child | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
2 children | −0.08*** | −0.06*** | −0.05** | −0.07 | −0.07 | −0.05 | −0.05* | −0.03 | −0.03 |
3 children | −0.16*** | −0.08*** | −0.06** | −0.18** | −0.14** | −0.13* | −0.03 | −0.01 | −0.01 |
Woman's age | |||||||||
<30 | 0.02 | 0.02 | −0.04 | −0.02 | 0.01 | −0.06 | 0.11*** | 0.11*** | 0.08*** |
>=30, <40 | 0.01 | 0.00 | −0.03 | −0.04 | −0.03 | −0.06 | 0.04* | 0.04** | 0.03 |
>=40, <50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
>=50, <60 | −0.05** | −0.06** | −0.02 | 0.03 | −0.01 | 0.02 | 0.00 | −0.01 | 0.00 |
>=60 | −0.09*** | −0.06* | 0.02 | −0.06 | −0.06 | −0.01 | −0.04 | −0.05 | −0.04 |
Missing | – | – | – | 0.60*** | 0.57*** | 0.68*** | 0.06 | 0.03 | 0.03 |
Age difference | |||||||||
Woman >=6y older | 0.06 | 0.07* | 0.03 | 0.08 | 0.14 | 0.10 | 0.07 | 0.09* | 0.08 |
Woman 3–5y older | 0.01 | 0.03 | 0.01 | 0.07 | 0.04 | 0.03 | 0.06 | 0.04 | 0.04 |
Same age ±2y | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Man 3–5y older | −0.02 | −0.04** | −0.03** | 0.01 | 0.01 | 0.01 | −0.01 | −0.01 | 0.00 |
Man >=6y older | 0.01 | 0.00 | −0.01 | 0.01 | −0.02 | −0.04 | 0.02 | 0.00 | 0.01 |
Missing | – | – | – | −0.41*** | −0.41*** | −0.40*** | −0.02 | −0.04 | 0.01 |
Stepfamily | |||||||||
No | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Yes | 0.12*** | 0.11*** | 0.07** | 0.20*** | 0.19*** | 0.16*** | 0.02 | 0.00 | −0.01 |
Socialization: mothers' employment status | |||||||||
Mother worked | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Mother did not work | 0.00 | 0.02 | 0.02* | 0.01 | 0.00 | 0.01 | −0.02 | 0.00 | 0.00 |
Missing | 0.01 | 0.03 | 0.03 | −0.03 | −0.01 | −0.01 | 0.02 | 0.01 | 0.01 |
Repondent's sex | |||||||||
Female | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Male | 0.12*** | 0.11*** | 0.11*** | 0.19*** | 0.15*** | 0.16*** | 0.24*** | 0.24*** | 0.25*** |
Interaction man's/woman's work status | |||||||||
Man working | |||||||||
Woman full-time | 0 | 0 | 0 | 0 | 0 | 0 | |||
Woman part-time | −0.07*** | −0.07*** | −0.07 | −0.07 | −0.13*** | −0.13*** | |||
Woman not working | −0.18*** | −0.18*** | −0.16*** | −0.17*** | −0.23*** | −0.21*** | |||
Man not working | |||||||||
Woman full-time | 0.25*** | 0.25*** | 0.42*** | 0.42*** | 0.36*** | 0.35*** | |||
Woman part-time | 0.13*** | 0.13*** | 0.18** | 0.18** | 0.09* | 0.09* | |||
Woman not working | −0.01 | −0.01 | 0.02 | 0.01 | −0.01 | 0.00 | |||
Woman's education | |||||||||
No vocational degree | 0 | 0 | 0 | 0 | 0 | 0 | |||
Vocational degree | 0.03* | 0.01 | 0.07 | 0.05 | 0.02 | 0.01 | |||
Tertiary | 0.11*** | 0.09*** | 0.19*** | 0.17** | 0.15*** | 0.13*** | |||
In education | 0.27*** | 0.25*** | 0.23** | 0.21** | 0.24*** | 0.20*** | |||
Missing | 0.16 | 0.15 | 0.29* | 0.30* | −0.05 | −0.01 | |||
Education difference | |||||||||
Same education | 0 | 0 | 0 | 0 | 0 | 0 | |||
Man higher | 0.02 | 0.02 | 0.10** | 0.10** | −0.01 | −0.01 | |||
Woman higher | −0.01 | −0.01 | 0.01 | 0.01 | 0.00 | 0.00 | |||
Missing | −0.10 | −0.13* | −0.30** | −0.23** | −0.05 | −0.03 | |||
Income ratio woman/man | |||||||||
0–<25% | 0 | 0 | 0 | 0 | 0 | 0 | |||
25%–<50% | 0.01 | 0.00 | −0.01 | −0.02 | −0.01 | −0.02 | |||
50%–<75% | 0.07*** | 0.06*** | 0.01 | 0.00 | −0.01 | −0.02 | |||
75%–<100% | 0.08*** | 0.07** | 0.08 | 0.05 | −0.07* | −0.07* | |||
100%–<125% | 0.10*** | 0.09*** | 0.11** | 0.08 | 0.04 | 0.03 | |||
>=125% | 0.12*** | 0.10*** | 0.10* | 0.09 | 0.13*** | 0.12*** | |||
Missing | −0.04 | −0.04 | 0.03 | 0.02 | −0.03 | −0.02 | |||
Gender attitudes: preschool child suffers if mother works | |||||||||
Strongly agree | 0 | 0 | 0 | ||||||
Agree | 0.02 | −0.16* | 0.04* | ||||||
Neither agree nor disagree | 0.00 | −0.17* | 0.08*** | ||||||
Disagree | 0.06*** | −0.09 | 0.08*** | ||||||
Strongly disagree | 0.05*** | −0.07 | 0.12*** | ||||||
Missing | 0.01 | 0.18 | 0.04 | ||||||
Marital status | |||||||||
Married | 0 | 0 | 0 | ||||||
Cohabiting | 0.04** | 0.06 | 0.12*** | ||||||
Partnership duration (years) | −0.0036*** | −0.0028 | 0.0010 | ||||||
Partnership duration missing | −0.27 | −0.14 | −0.09** | ||||||
Constant | −0.35*** | −0.46*** | −0.43*** | −0.37*** | −0.55*** | −0.38*** | −0.53*** | −0.57*** | −0.66*** |
N | 5450 | 5450 | 5450 | 1129 | 1129 | 1129 | 4789 | 4789 | 4789 |
. | France . | Eastern Germany . | Western Germany . | ||||||
---|---|---|---|---|---|---|---|---|---|
. | Model 1 . | Model 2 . | Model 3 . | Model 1 . | Model 2 . | Model 3 . | Model 1 . | Model 2 . | Model 3 . |
Age of the youngest child | |||||||||
No children | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Age 0–5 | −0.11*** | −0.05** | −0.05* | −0.13* | −0.05 | −0.04 | −0.21*** | −0.08** | −0.05 |
Age 6–13 | −0.16*** | −0.12*** | −0.11*** | −0.10 | −0.05 | −0.04 | −0.24*** | −0.13*** | −0.11*** |
Age 14–17 | −0.21*** | −0.17*** | −0.14*** | −0.16** | −0.12* | −0.11 | −0.15*** | −0.05 | −0.03 |
Age 18+ | −0.27*** | −0.22*** | −0.19*** | −0.29*** | −0.24*** | −0.22*** | −0.09*** | −0.03 | −0.02 |
Non-resident children only | −0.19*** | −0.15*** | −0.13*** | −0.06 | −0.02 | −0.01 | −0.05* | 0.00 | 0.00 |
Number of children | |||||||||
1 child | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
2 children | −0.08*** | −0.06*** | −0.05** | −0.07 | −0.07 | −0.05 | −0.05* | −0.03 | −0.03 |
3 children | −0.16*** | −0.08*** | −0.06** | −0.18** | −0.14** | −0.13* | −0.03 | −0.01 | −0.01 |
Woman's age | |||||||||
<30 | 0.02 | 0.02 | −0.04 | −0.02 | 0.01 | −0.06 | 0.11*** | 0.11*** | 0.08*** |
>=30, <40 | 0.01 | 0.00 | −0.03 | −0.04 | −0.03 | −0.06 | 0.04* | 0.04** | 0.03 |
>=40, <50 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
>=50, <60 | −0.05** | −0.06** | −0.02 | 0.03 | −0.01 | 0.02 | 0.00 | −0.01 | 0.00 |
>=60 | −0.09*** | −0.06* | 0.02 | −0.06 | −0.06 | −0.01 | −0.04 | −0.05 | −0.04 |
Missing | – | – | – | 0.60*** | 0.57*** | 0.68*** | 0.06 | 0.03 | 0.03 |
Age difference | |||||||||
Woman >=6y older | 0.06 | 0.07* | 0.03 | 0.08 | 0.14 | 0.10 | 0.07 | 0.09* | 0.08 |
Woman 3–5y older | 0.01 | 0.03 | 0.01 | 0.07 | 0.04 | 0.03 | 0.06 | 0.04 | 0.04 |
Same age ±2y | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Man 3–5y older | −0.02 | −0.04** | −0.03** | 0.01 | 0.01 | 0.01 | −0.01 | −0.01 | 0.00 |
Man >=6y older | 0.01 | 0.00 | −0.01 | 0.01 | −0.02 | −0.04 | 0.02 | 0.00 | 0.01 |
Missing | – | – | – | −0.41*** | −0.41*** | −0.40*** | −0.02 | −0.04 | 0.01 |
Stepfamily | |||||||||
No | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Yes | 0.12*** | 0.11*** | 0.07** | 0.20*** | 0.19*** | 0.16*** | 0.02 | 0.00 | −0.01 |
Socialization: mothers' employment status | |||||||||
Mother worked | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Mother did not work | 0.00 | 0.02 | 0.02* | 0.01 | 0.00 | 0.01 | −0.02 | 0.00 | 0.00 |
Missing | 0.01 | 0.03 | 0.03 | −0.03 | −0.01 | −0.01 | 0.02 | 0.01 | 0.01 |
Repondent's sex | |||||||||
Female | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |
Male | 0.12*** | 0.11*** | 0.11*** | 0.19*** | 0.15*** | 0.16*** | 0.24*** | 0.24*** | 0.25*** |
Interaction man's/woman's work status | |||||||||
Man working | |||||||||
Woman full-time | 0 | 0 | 0 | 0 | 0 | 0 | |||
Woman part-time | −0.07*** | −0.07*** | −0.07 | −0.07 | −0.13*** | −0.13*** | |||
Woman not working | −0.18*** | −0.18*** | −0.16*** | −0.17*** | −0.23*** | −0.21*** | |||
Man not working | |||||||||
Woman full-time | 0.25*** | 0.25*** | 0.42*** | 0.42*** | 0.36*** | 0.35*** | |||
Woman part-time | 0.13*** | 0.13*** | 0.18** | 0.18** | 0.09* | 0.09* | |||
Woman not working | −0.01 | −0.01 | 0.02 | 0.01 | −0.01 | 0.00 | |||
Woman's education | |||||||||
No vocational degree | 0 | 0 | 0 | 0 | 0 | 0 | |||
Vocational degree | 0.03* | 0.01 | 0.07 | 0.05 | 0.02 | 0.01 | |||
Tertiary | 0.11*** | 0.09*** | 0.19*** | 0.17** | 0.15*** | 0.13*** | |||
In education | 0.27*** | 0.25*** | 0.23** | 0.21** | 0.24*** | 0.20*** | |||
Missing | 0.16 | 0.15 | 0.29* | 0.30* | −0.05 | −0.01 | |||
Education difference | |||||||||
Same education | 0 | 0 | 0 | 0 | 0 | 0 | |||
Man higher | 0.02 | 0.02 | 0.10** | 0.10** | −0.01 | −0.01 | |||
Woman higher | −0.01 | −0.01 | 0.01 | 0.01 | 0.00 | 0.00 | |||
Missing | −0.10 | −0.13* | −0.30** | −0.23** | −0.05 | −0.03 | |||
Income ratio woman/man | |||||||||
0–<25% | 0 | 0 | 0 | 0 | 0 | 0 | |||
25%–<50% | 0.01 | 0.00 | −0.01 | −0.02 | −0.01 | −0.02 | |||
50%–<75% | 0.07*** | 0.06*** | 0.01 | 0.00 | −0.01 | −0.02 | |||
75%–<100% | 0.08*** | 0.07** | 0.08 | 0.05 | −0.07* | −0.07* | |||
100%–<125% | 0.10*** | 0.09*** | 0.11** | 0.08 | 0.04 | 0.03 | |||
>=125% | 0.12*** | 0.10*** | 0.10* | 0.09 | 0.13*** | 0.12*** | |||
Missing | −0.04 | −0.04 | 0.03 | 0.02 | −0.03 | −0.02 | |||
Gender attitudes: preschool child suffers if mother works | |||||||||
Strongly agree | 0 | 0 | 0 | ||||||
Agree | 0.02 | −0.16* | 0.04* | ||||||
Neither agree nor disagree | 0.00 | −0.17* | 0.08*** | ||||||
Disagree | 0.06*** | −0.09 | 0.08*** | ||||||
Strongly disagree | 0.05*** | −0.07 | 0.12*** | ||||||
Missing | 0.01 | 0.18 | 0.04 | ||||||
Marital status | |||||||||
Married | 0 | 0 | 0 | ||||||
Cohabiting | 0.04** | 0.06 | 0.12*** | ||||||
Partnership duration (years) | −0.0036*** | −0.0028 | 0.0010 | ||||||
Partnership duration missing | −0.27 | −0.14 | −0.09** | ||||||
Constant | −0.35*** | −0.46*** | −0.43*** | −0.37*** | −0.55*** | −0.38*** | −0.53*** | −0.57*** | −0.66*** |
N | 5450 | 5450 | 5450 | 1129 | 1129 | 1129 | 4789 | 4789 | 4789 |
p<.1; **p<.05; ***p<.01.
Source: GGS 2005, own calculations.
In general, we had expected the division of household tasks to become more egalitarian again for couples with older children, and to re-approach the level for childless couples. We had expected the return to a more egalitarian division of labor to be swifter in France and in eastern Germany than in western Germany, where mothers' employment breaks tend to be longer. For western Germany, we do find the general pattern that we had expected. The division of everyday household tasks is more egalitarian for couples with a youngest child aged 14–17 than for couples with younger children. Couples move yet further in the direction of an egalitarian division of labor when their children are aged 18 or over or have all left the parental home.
Overall, the division of housework is clearly more egalitarian in France and eastern Germany than in western Germany, and the difference is greatest among those with young children. This fits our expectations given higher maternal employment rates in the former two countries and regions. The shape of the effect of the age of the youngest child is however not what we expected for France and eastern Germany. Instead of a quick return to a more egalitarian division of labor, parents' division of household tasks becomes continuously less egalitarian for parents with older children. Children have been held to have a traditionalizing impact on couples' division of housework due to norms of domesticity associated with motherhood, as well as via a stronger routinization of housework adopted by parents in order to cope with the increased workload (Kaufmann 1999; Pfau-Effinger 2004; van der Lippe et al. 2010). Our results indicate that the traditionalizing impact of parenthood unfolds slowly as children grow older. It is likely that the routinization of housework described by Kaufmann (1999) takes time to build up, and gradually becomes more pronounced. According to Kaufmann, when parents introduce more routines and standardize the allocation of housework, they often resort to traditional gender role patterns as a point of orientation.
An alternative explanation may be that fathers' contributions to housework are particularly high when children are young and the household's total workload is high. As children grow older and the household's total workload is reduced, fathers may then withdraw their contributions, leaving the remaining housework largely to women (Brugeilles and Sebille 2009). On the basis of the data used for our study, however, we cannot conclusively determine to what extent this second explanation holds. Our data provide information on partners' relative housework contributions, but not on the absolute amount of time spent on housework.
Previous research has shown that couples' division of housework also becomes less egalitarian at longer partnership durations (Schulz and Blossfeld 2006; Wengler et al. 2009). We include partnership duration in our third model. We find a significant effect of partnership duration only for France, which does not, however, explain the effect of the age of the youngest child.
Finally, we find that when children leave the parental home, this triggers a move towards a more egalitarian division of housework in all three countries or regions. This could be related to parents' retirement. However, we continue to find this effect even when controlling for employment status. Possibly, the traditionalizing impact of parenthood is at least partially reversible when children leave home. The reduction in the total household workload when children leave home should make it less important for parents to adhere to the strict housework routines they introduced sometime after the birth of their children (Kaufmann 1999), opening a pathway for more flexibility in their allocation of household tasks again.
In our second model, we include variables for men's and women's employment status, education, and the couple's income ratio. Including these variables serves to test whether changes in couples' division of housework are associated with changes in women's employment status or income. Adding these variables however does very little to explain the effect of the age of the youngest child in France and in eastern Germany. In couples with children aged over 14, men's contribution to everyday household tasks is still significantly lower than for childless couples. This provides further support for the idea that the changes in couples' division of household tasks by age of the youngest child are not driven by changes in the allocation of women's time between employment and household work, but may indicate a long-term traditionalizing influence of norms regarding parenthood roles.
In western Germany, however, including these variables explains a larger part of the difference between childless couples, couples with younger children, and couples with older children. This seems to indicate that the changes in the division of housework with the age of the youngest child are indeed driven by reallocations of women's time between employment and housework in western Germany. This does not rule out that a long-term traditionalizing impact of parenthood, as suspected for France and eastern Germany, exists for western Germany as well. The effect of changes in mothers' employment status may however be so strong in western Germany that they dominate over any further trends associated with the age of the youngest child.
In our third model, we include marital status and gender attitudes. As expected, the division of everyday household tasks is more egalitarian for cohabiting than for married couples. People who believe that preschool children suffer if their mother works also tend to be in couples with a less egalitarian division of labor. We had initially included several further indicators of gender attitudes. This indicator was however the only one that proved to have a significant effect. Including variables for marital status and gender attitudes only slightly changes the parenthood effect.
Most of the control variables included in our models have effects in the expected directions. We will only briefly discuss the effects of a few selected control variables due to space limitations. A greater number of children, for instance, leads to a less egalitarian division of housework, particularly in France and in eastern Germany. This may be related to lower maternal employment rates as well as to stronger normative parental role expectations. In France and in eastern Germany, the division of housework is also clearly more egalitarian in stepfamilies. As discussed in the theoretical background section, fewer normative expectations with regard to the division of housework may exist for second partnerships. Our results indicate that this applies for stepfamilies as well, and are thus in line with findings by Snoeckx et al. (2008). It is however possible that at the same time, the division of childcare, which we do not analyze in our study, is less egalitarian in stepfamilies. Our findings for effects of combinations of partners' employment statuses support theories of time availability. In couples where the woman is working part-time and especially in those where she is not employed, she contributes more to the housework than in couples where the woman is employed full-time. In couples where the man is not employed, he contributes more to the housework than where he is employed. This finding supports arguments by Sullivan (2011) who has contested the gender display hypothesis. The gender display hypothesis had contrarily held that non-employed men perform less housework, thereby displaying their gender, in order to correct for their deviation from traditional gender role expectations. A test of the influence of relative resources is provided by including a measure of couples' relative incomes. Relative income effects are particularly clear in France. Here we can see that the higher the woman's income in relation to the man's, the more the man contributes to the housework. There is also a tendency in this direction in eastern Germany. In western Germany, this effect is not quite as clear. This might have to do with the comparatively small proportion of women whose earnings are close to their partner's in western Germany.
7. Conclusion
In this study, we investigated how couples' division of household tasks is related to the age of the youngest child, comparing France, eastern Germany, and western Germany. Our findings are that the division of housework is much less egalitarian among parents of very young children in western Germany than in France or eastern Germany. For couples with older children, the division of housework gradually converges in all three countries and regions.
Thus, it seems that in all three countries and regions, parenthood has a traditionalizing impact on the division of housework. In western Germany, changes in couples' division of housework are quite abrupt for those with young children compared to the childless. In France and eastern Germany, the change is weaker and more gradual. It appears that while couples in France and eastern Germany initially attempt to sustain an egalitarian division of housework upon becoming parents, they eventually at least partly yield to traditional role expectations associated with parenthood. A likely explanation is that parents increasingly introduce housework routines to cope with the greater workload, and that this opens a pathway for the influence of traditional gender role expectations (Kaufmann 1994, 1999).
The different patterns found for the division of housework by age of the youngest child are likely to reflect cultural and institutional differences between all three countries and regions. Eastern Germany and France have been described as exhibiting a stronger societal orientation towards the dual earner model of the family (Lewis 1992; Pfau-Effinger and Smidt 2011). Western Germany, on the other hand, has been described as being culturally and institutionally more strongly oriented towards the male breadwinner model of the family (Pfau-Effinger 2004). In western Germany, the childcare infrastructure is much less extensive than in France or in eastern Germany. For couples with young children in western Germany, options for both partners to be employed are thus limited. Many mothers in western Germany therefore decide to take long breaks from employment. For those with older children, maternal employment rates rise again. Our findings show that the division of housework also becomes more egalitarian again for those with older children, and becomes more similar to that in France or eastern Germany. We show that the U-shaped development in the division of housework by age of the youngest child in western Germany can to a large part be accounted for by changes in women's employment status.
In part, parents' decisions on the division of housework may however also directly follow normative cultural ideals regarding the division of housework, irrespective of their employment status. This could explain why, even among dual earner couples, the division of housework is still somewhat less egalitarian for parents of young children in western Germany than in France or in eastern Germany.
Footnotes
International comparative studies on gender and motherhood wage gaps usually do not differentiate between eastern and western Germany. On account of the greater population size in western Germany, findings for Germany are likely to more closely reflect the situation in western Germany.
The GGS data used for this study were obtained from the GGP Data Archive. For a description of the survey instruments, see Generations & Gender Programme: Survey Instruments. United Nations (2005).
These robustness checks are not included in the paper in order to save space. However, they are available on request.
Men as well as women may overstate their own housework contributions (Batalova and Cohen 2002; Baxter 2005). We have therefore controlled for the respondent's gender in each model in order to correct for any potential biases that might be caused by uneven distributions of men and women across the samples.
References
Cordula Zabel is a senior research scientist at the Institute for Employment Research (IAB), in Nuremberg, Germany. Previously, she was a Ph.D. student and a Post-Doc at the Max Planck Institute for Demographic Research in Rostock, Germany. Her research interests are in the fields of household dynamics, labor market transitions, and social policy.
Valerie K. Heintz-Martin received her Ph.D. from McGill University in Montreal, Canada. Before joining the German Youth Institute (DJI) in Munich in 2011, she was an Assistant Professor at Bamberg University, Germany. Her research interests are in the areas of fertility, stepfamilies, international comparisons of family types, and life course research.