Men’s Employment Hours and Time on Domestic Chores in European Countries

Abstract

Over the past decades, the gender gap in housework has become smaller and scholars have called on changing structural conditions and on the diffusion of egalitarian gender roles to explain why. In particular, women’s presence in the public sphere is found to be associated with a more egalitarian division of chores between partners. However, despite the large presence of women in the public sphere in many countries, the gender gap in housework has not disappeared. This article asks whether the widespread presence of men in the public sphere is slowing the diffusion of more egalitarian practices of housework division. Using multilevel models on European Social Survey data (2010), the article shows that in countries where men work long standard hours, women perform relatively more housework and men relatively less, highlighting the importance of men’s aggregate behavior in explaining partners’ relative time on housework.

Keywords

men’s housework time relative time on housework intrahousehold allocation multilevel models Europe

Introduction

Women and men’s different allocation of time to domestic chores has been the subject of much comparative research (among many others: Breen & Cooke, 2005; Geist, 2005; Hook, 2010; Ruppanner, 2010, 2012; Treas & Drobnič, 2010; Treas & Tai, 2012). Today, housework is still mostly a “woman’s thing,” but studies on several Western countries show that over the past 60 years, women have reduced the time they devote to domestic chores, men have increased it, and the gender gap in housework has become narrower (Gershuny, 2000; Hook, 2010; Sayer, 2010).

One of the reasons for the narrowing of the gender gap in housework can be sought in the diffusion of egalitarian values linked with the Second Demographic Transition (Lesthaeghe, 2010). Indeed, several contributions have shown that egalitarian values translate into a more equal division of chores both at the individual and at the societal level, reflecting the diffusion of gender egalitarian norms of behavior (e.g., Yodanis, 2005). Other scholars, instead, have emphasized the structural changes in the relationship between genders and have called on these to explain the unequal division of chores. Among the structural characteristics considered in several of these studies, the presence of women in the public sphere has been found to favor equality in the division of chores within couples (Fuwa, 2004; Knudsen & Wærness, 2008). In fact, as Jennifer Hook (2006) pointed out, “In contexts where women are more involved in the public sphere, men are more involved in the private sphere, not necessarily because of household bargaining or other household-level processes, but because of societal shifts in gendered behavior” (p. 643).

Regardless of the advances in the individualization processes in contemporary societies, and of the increasing presence of women in the public sphere, the inequality in the division of housework between partners persists (Röhler & Huinink, 2010). However, just as the presence of women in the public sphere favors a more egalitarian division of chores, other structural factors may be hindering the diffusion of more gender-equal practices of housework division.

This article asks whether another structural feature, the extensive presence of men in the public sphere, is stalling the process that has led to a more equal division of chores within households. On the one hand, long working hours could represent a barrier to men spending more time at home and sharing domestic chores (Gerson, 2010). On the other hand, just as a large presence of women in the public sphere may signal that women and men should share housework more evenly, the extensive presence of men in the public sphere may have the opposite effect and reinforce the traditional gendered division of domestic labor. Using a multilevel research strategy on the fifth wave of the European Social Survey (2010), I analyze whether long average working hours for men are associated with a more unequal allocation of time to domestic chores between partners. The analyses show that women spend more relative time on domestic chores in countries where men on average work longer hours. Furthermore, the result is robust to the inclusion of two macro-level variables that control for the presence of women in the public sphere. This indicates that men’s working hours are very important in perpetuating an unequal division of domestic labor within couples. In particular in the case of women, men’s working hours are the strongest predictor of an unequal allocation of time to housework.

The article has the following structure. In the next section, I present the theoretical backbone of the contribution: I discuss the links between individual behavior and structural characteristics and I outline my hypotheses. Then, I introduce the historical perspectives that explain the allocation of time to domestic chores at the individual and couple levels. In the following section, I introduce the data, sample, variables, and methods, while in the penultimate section, I present the results. The final section discusses and concludes.

Adapting to the “Proper Practice” in the Division of Domestic Chores: Back and Forth Between Individual and Country Characteristics

A well-established line of research has shown that the division of domestic chores within couples largely depends on the partners’ characteristics. Recent studies, however, have found that country characteristics are also associated with the way partnered women and men share domestic chores (Treas & Drobnič, 2010).

It is not surprising that individuals’ behavior is guided by the characteristics of the environment they are embedded in. Referring to Burt’s (1987) work on structural equivalence, Nazio and Blossfeld (2003) nicely phrase that individuals behave not only according to their personal traits, but also according to the “perception of the practice proper for an individual of her position within the social structure” (p. 52). In terms of housework, there are many structural features that can signal the “proper practice.” Hook (2006) provides an example regarding female employment: “as women’s labor force participation affects more men, the bar is set higher when men make social comparisons, creating an across-the-board change in how men “do gender” [ . . . ]” (p. 643). Indeed, several authors have looked at how female labor force participation and the degree of women’s empowerment in societies are related to the division of domestic chores in households (Batalova & Cohen, 2002; Fuwa, 2004; Knudsen & Wærness, 2008). The results have been somewhat inconsistent across studies. Batalova and Cohen (2002), Fuwa (2004), and Ruppanner (2010), on the one hand, show that in countries where women are more empowered, the division of chores is more equal. Knudsen and Wærness (2008), on the other hand, find no direct effect of women’s empowerment on women’s share of domestic chores. The results for female labor force participation are also not straightforward. Hook (2006) finds that men perform more housework in countries where women are overall present in the labor market. In a successive study, the author shows that the presence of married women in the work force increases the time that men spend on cooking and on housework (Hook, 2010). Treas and Tai (2012) also show that housework is shared more evenly in countries that have a high legacy of maternal employment. This could suggest that female labor force participation has an equalizing effect on the division of housework. Fuwa (2004), however, directly tests this hypothesis and finds no effect of female labor force participation on the division of chores within couples. Despite not being fully consistent between studies, these results point toward a relation between women’s presence in the public sphere and a more egalitarian division of chores within couples.

The change in individual behavior that comes with the adoption of the “proper practice,” regardless of personal characteristics, has been referred to as an equalization process that goes beyond the individual-level bargaining within the household (Hook, 2006). Despite the ongoing equalization process, however, the time women and men spend on housework has not converged. To account for this, in this article I ask what is the role of men’s aggregate behavior in the equalization process of the division of housework. Research has already shown that longer working hours for men result in more absolute time spent on domestic chores for women (Hook, 2010). I argue that the same mechanism may be at work in the relative amount of time on chores: long standard working hours, in fact, may signal that the right place for a man is the workplace. Hence, normative standards may encourage men not to share household chores in order to live up to the expectations of the traditional gendered division of labor (Coleman, 1991), regardless of their own working hours. Likewise, women might feel social pressure to perform more than their share of housework when they live in “traditional” contexts (Geist, 2005), for example, where men on average spend a lot of their time at the work place. Furthermore, beyond signaling proper practices, men’s long standard working hours may result in aggregate time binds that could make it difficult for them to increase their contribution to housework, regardless of their own work time. Following these lines of reasoning, I formulate the first hypothesis:

Hypothesis 1: In countries where men’s standard working hours are longer, the allocation of time to domestic chores between partners is less equal.

Previous research has placed much emphasis on the importance of women’s aggregate behavior in the equalization process and, as I have shown, the results have been somewhat inconsistent across studies. Therefore, it is important to test whether the presence of women in the public sphere is associated with a more egalitarian division of chores. To this end, I hypothesize a positive association between, on the one hand, female labor force participation¹ and women’s empowerment and, on the other hand, a more equal allocation of time to domestic chores within couples:

Hypothesis 2: In countries where female labor force participation rates are high or women are more empowered, the allocation of time to domestic chores between partners is more equal.

Finally, which of the three considered structural characteristics has the strongest association with the outcome? If men’s presence in the public sphere is stalling the equalization process, then we would expect women’s presence in the public sphere to have a negligible association with the outcome. Thus, we can hypothesize the following:

Hypothesis 3a: Once men’s average working hours are controlled for, the association between the relative time spent on housework and the presence of women in the public sphere shall be weak or absent.

On the contrary, men’s hours in the public sphere could have nothing to do with the slowdown of the equalization process. In this case, we would expect men’s average working hours to lose statistical power once the variables indicating the presence of women in the public sphere are included in the model. Therefore, the competing hypothesis is as follows:

Hypothesis 3b: Once the presence of women in the public sphere is controlled for, the association between the relative time spent on housework and men’s average working hours shall be weak or absent.

Additional Considerations on the Division of Domestic Chores

Since the 1960s, three theoretical perspectives have emerged to explain what individual traits are associated with the division of household tasks between partners. I summarize these theories to define what control variables need to be included at the individual- and household-level of the analyses. The time availability perspective (Hiller, 1984) explains the division of household labor as the result of a rational process, where household tasks are allocated according to the available time of both partners. The main indicators used to measure time constraints are the employment status and schedules of the partners and the presence and number of children in the household. Husbands do more domestic chores if their wives are employed outside the home (Davis & Greenstein, 2004), and both partners devote more time to domestic chores the more hours they spend at home (Geist, 2005; Presser, 1994). Similarly, parenthood is found to increase the domestic work load and also to strengthen a traditional division of labor, both in cross-sectional (Craig & Mullan, 2010; Dribe & Stanfors, 2011) and in longitudinal studies (Kühhirt, 2012; Schober, 2013).

According to the relative resources perspective, instead, the division of domestic chores is determined by the level of relative resources each partner brings to the relationship (Blood & Wolfe, 1960). Research has found that the smaller the wage gap between husband and wife, the more household tasks are equally divided (Blair & Lichter, 1991; Greenstein, 2000), while men’s relatively higher income leads women to do more housework (Geist, 2005). Resources have also been operationalized through educational level. Husbands perform more housework when their wives’ education is equal to their own or exceeds it (Blair & Lichter, 1991; Davis & Greenstein, 2004). However, additional education is not just linked with financial resources, but also with more egalitarian attitudes. Hence, in this case, the higher the husband’s educational level, the more likely is his participation to household tasks (South & Spitze, 1994).

The social construction of gender perspective (Brines, 1994; Ferree, 1990) argues that the division of housework is a result of how men and women display their social roles and “produce gender” (Connell, 1985; Hochschild, 1989; West & Zimmerman, 1987). For example, traditional women are more likely to consider “fair” an unequal division of chores because it matches with the normative standards they embrace (Lavee & Katz, 2002). Conversely, men with more traditional gender ideologies do less housework than those who are more modern because they might not want to appear “unmanly” (Brayfield, 1992; Huber & Spitze, 1983).

Research Strategy

Data and Sample

The data for the analyses are drawn from the fifth wave of the European Social Survey (2010, ESS hereafter). The ESS is a biennial survey that involves over 30 nations and has been carried out since 2002/2003. Its core questionnaire captures Europeans’ attitudes and values on a vast number of topics. In each survey, the core questionnaire is integrated with a different rotating module. The fifth wave rotating module is titled “Family, Work and Well-Being” and is an improved repetition of a similar one included in the second wave. The advantage of this survey is that it collects information on how much time the respondent and the partner spend on domestic chores and therefore is one of the few data sets that allows investigating the relative time spent on housework by partners in a large number of countries. Most large-scale cross-national data sets, instead, such as the European Value Study, the World Value Survey, even the Multinational Time Use Study, miss the appropriate comparable information to study the relative time couples devote to domestic chores.² There are, however, two important drawbacks that need to be highlighted. First, there has been evidence that time use reported through direct questions—as opposed to data collected through diaries—tends to be biased, as respondents overestimate the time devoted to frequent activities and underestimate the time devote to infrequent ones (Juster & Stafford, 1991; Marini & Shelton, 1993; Niemi, 1993). A second problem is that the partner is not interviewed and the respondent reports own time on housework and the partner’s time on housework. Previous cross-national studies have used these measures of time, problematic as they may be (e.g., Knudsen & Wærness, 2008). The findings, however, can be interpreted only as based on the perception of time used, and not actual time used, by the respondents and their partners.

I select individuals who are between 20 and 65 years old, living with a partner, and residing in one of the following 23 countries: Belgium (BE), Bulgaria (BG), Switzerland (CH), Cyprus (CY), Czech Republic (CZ), Germany (DE), Denmark (DK), Estonia (EE), Spain (ES), Finland (FI), France (FR), Great Britain (GB), Greece (GR), Croatia (HR), Hungary (HU), Ireland (IE), the Netherlands (NL), Norway (NO), Poland (PL), Portugal (PT), Sweden (SE), Slovenia (SI), and Slovakia (SK).

Dependent Variable

In the ESS questionnaire, housework is defined as “things done around the home, such as cooking, washing, cleaning, care of clothes, shopping, maintenance of property, but not including child care or leisure activities” and is measured in hours per week.³ In order to capture the relative amount of time each partner spends on housework, similarly to Knudsen and Wærness (2008), I build a dependent variable that measures the reported time each partner spends on domestic chores relative to the reported total time partners spend on such activities. The dependent variable ranges from 0 to 1, where 0 stands for the partner doing all the housework, and 1 stands for the respondent doing all the housework.⁴

Independent Variables

To test whether men’s standard working hours are associated with a more unequal allocation of time to domestic chores between partners, I include at the country-level, the average hours worked by men per week in their main job (International Labour Organization, 2010). Women’s presence in the public sphere is operationalized using the employment rate of women aged 15 to 64 years old (Eurostat, 2012) and the Gender Empowerment Measure (GEM; United Nations Development Programme, 2007), as done in previous research (e.g., Fuwa, 2004; Knudsen & Wærness, 2008). The GEM is a measure of women’s empowerment and is built on three dimensions: (a) political participation and decision making; (b) economic participation and decision making, and (c) power over economic resources. Political participation and decision making is measured using the female and male shares of parliamentary seats, while economic participation and decision making is calculated by combining female and male shares of positions as legislators, senior officials, and managers, and female and male shares of professional and technical positions. The third dimension, power over economic resources, is calculated using female- and male-estimated earned income. All three macro-level variables are measured circa 2010 and are included in the models in standardized form in order for the estimates to be comparable.

As the dependent variable involves the activities of two household members, I include household-level predictors along with individual-level covariates. The baseline model includes age (centered at the grand mean), marital status (married vs. not legally married), number of household members, and total time spent on housework by the partners. This variable allows controlling for the fact that the amount of housework may influence the way partners allocate time to domestic chores. I then add the so-called time constraints variables: employment status (employed vs. not employed), hours in employment per week (coded 0 for those out of employment⁵), employment status of the partner (employed vs. not employed), hours in employment per week of the partner, and presence of children younger than 18 years old in the household. The models also control for the partners’ economic and educational resources. The first are included using an indicator that measures how large a proportion of the household income the respondent provides. Respondents were asked whether they earn none of the household income (reference category), a very small part, under a half, about half, over a half, a very large part, or all. To measure relative educational resources, I build a semicompound variable that includes five combinations⁶ of the respondent’s and the partner’s levels of education⁷: both low⁸ (reference category); both medium; both high; partner is more educated than the respondent; and partner is less educated than the respondent. Attitudes toward gender roles are controlled for using the answers from the following question: “women should be prepared to cut down on paid work for the sake of the family.” The responses range from 1 (agree strongly) to 5 (disagree strongly).⁹ Note that the respondents’ attitudes and the reported time on housework are measured at the same time. It is therefore impossible to assess whether egalitarian gender attitudes foster an egalitarian division of chores or the opposite, and this limitation must be kept in mind when interpreting the results. Table 1 reports the grand means and standard deviations for all the variables included in the models. Table 2 reports means and standard deviations for the macro-level variables along with sample sizes by country and gender.

Table 1.

Descriptive Statistics.

	Women		Men
Individual- and household-level indicators	Mean	SD	Mean	SD
Calculated proportion of housework (min 0–max 1)	0.72	0.19	0.32	0.19
Reported hours of housework per week (min 0–max 84)	20.46	13.59	8.99	8.51
Reported partner’s hours of housework per week (min 0–max 84)	8.12	8.84	19.96	13.86
Calculated total hours of housework per week (min 0–max 168)	28.58	17.96	28.95	17.71
Age (min 20–max 65)	45.13	11.7	46.63	11.41
Number of household members (min 2–max 14)	3.15	1.17	3.11	1.11
Reported hours of employment per week (min 0–max 84)	21.34	20.15	33.57	21.26
Reported partner’s hours of employment per week (min 0–max 84)	32.24	20.68	23.27	19.18
Proportion of household income provided (min 1 [none]–max 7 [all])	3.21	1.39	4.90	1.35
Value orientation: “Women should be prepared to cut down on paid work for the sake of the family” (min 1 [agree strongly]–max 5 [disagree strongly]	2.99	1.20	3.02	1.15
	Proportions
Married vs. not married	0.81		0.79
Employed vs. not employed	0.58		0.76
Partner employed vs. partner not employed	0.75		0.65
Presence of children younger than 18 years in the household (0 [no]–1 [yes])	0.61		0.60
Relative education
Both partners low educated	0.14		0.13
Both partners medium educated	0.30		0.31
Both partners high educated	0.19		0.20
Partner more educated	0.17		0.17
Partners less educated	0.19		0.19
N	8,778		7,596
Country-level indicators	Mean		SD
Employed men’s weekly paid hours (min 37–max 43)	40		1.50
Percentage of women, 15-64 years, in labor force (min 0.48–max 0.73)	0.60		0.07
Gender Empowerment Measure (min 0.56–max 0.91)	0.72		0.11

Note. min = minimum; max = maximum.

Table 2.

Descriptive Statistics: Dependent Variable by Country and Gender and Macro-Level Measures by Country.

	Reported proportion of housework						Country-level measures
	Women			Men			Hours of employment/week (men)	Women 15-64 years old in workforce (%)	Gender Empowerment Measure
	Mean	SD	N	Mean	SD	N	Hours of employment/week (men)	Women 15-64 years old in workforce (%)	Gender Empowerment Measure
Belgium	0.72	0.19	431	0.30	0.20	365	39	0.56	0.85
Bulgaria	0.73	0.16	427	0.29	0.16	289	40	0.56	0.60
Switzerland	0.74	0.19	317	0.27	0.19	335	43	0.72	0.66
Cyprus	0.83	0.19	201	0.26	0.23	154	41	0.63	0.58
Czech Republic	0.71	0.16	402	0.30	0.17	399	40	0.56	0.63
Germany	0.73	0.18	644	0.29	0.18	616	40	0.66	0.83
Denmark	0.64	0.18	386	0.37	0.18	411	38	0.71	0.87
Estonia	0.65	0.17	381	0.37	0.16	274	39	0.60	0.64
Spain	0.73	0.21	404	0.27	0.22	380	40	0.52	0.79
Finland	0.63	0.18	416	0.38	0.17	432	38	0.67	0.89
France	0.72	0.20	377	0.31	0.21	342	39	0.60	0.72
Great Britain	0.73	0.21	433	0.33	0.20	371	41	0.64	0.78
Greece	0.85	0.18	608	0.20	0.22	349	43	0.48	0.62
Croatia	0.76	0.18	252	0.30	0.20	221	41	0.49	0.61
Hungary	0.73	0.18	325	0.29	0.18	292	40	0.50	0.57
Ireland	0.75	0.20	545	0.38	0.21	376	39	0.56	0.70
The Netherlands	0.72	0.19	456	0.32	0.19	394	37	0.69	0.86
Norway	0.67	0.18	380	0.37	0.16	424	38	0.73	0.91
Poland	0.69	0.19	304	0.32	0.17	325	41	0.53	0.61
Portugal	0.83	0.18	270	0.30	0.21	176	40	0.61	0.69
Sweden	0.63	0.16	336	0.43	0.16	315	38	0.70	0.90
Slovenia	0.71	0.18	283	0.33	0.21	232	40	0.63	0.61
Slovakia	0.64	0.18	200	0.39	0.20	124	40	0.52	0.63

Method

The hierarchical structure of the data calls for the use of multilevel models, which control for clustering within groups/levels (Hox, 2010). I use two-level random intercepts models with individuals nested in countries. I run the models separately for men and women, as the literature has shown that there are very large gender differences in the time allocated to domestic chores, both in absolute and in relative terms. Running models separately allows seeing immediately whether there are gender differences in the association between individual- and household-level traits and the reported shares of housework.¹⁰ Given that the dependent variable is a proportion and ranges from 0 to 1, rather than modeling the outcome as continuous, I follow the approach suggested by Hox (2010) and by McDowell and Cox (2004) and use a generalized linear model with a logit link and the binomial family. While this approach has the advantage of producing predicted values between 0 and 1, the raw coefficients are not interpretable as the response proportions. To grasp a better understanding of the magnitude of the coefficients, I calculate predictions of the outcome¹¹ and display graphically the results for the macro-level variables.

Results

Descriptive Analysis

As expected, women report performing more than half of the domestic chores even in countries where the allocation of time to housework between partners is most equal. As can be seen in Table 1, which reports grand means and standard deviations for the overall sample, women on average perform more than 70% of the domestic workload. Table 2 shows the dependent variable by country and gender and allows more detail in the relative time on chores in different contexts. The most unequal division of chores is found in Greece, where women perform 85% of the relative time on domestic work. Women in Cyprus and Portugal follow with 83%. Sweden and Finland are the countries where women do the smallest share of housework, 63%. Moving to a more accurate description of the allocation of time to housework in context, in the following sections I discuss the results from the multilevel models, focusing first on the macro-level variables and then on the individual-level predictors.

Country-Level Predictors

Table 3 reports, separately by gender, the log-odds for five multilevel random intercepts models. Model 1 includes only the individual-level predictors. Models 2, 3, and 4 add, respectively, men’s average working hours, female labor force participation, and the GEM. Finally, Model 5 includes all three macro-level variables at once. Each model also controls for all the individual-level predictors but, for space limitations, Table 3 reports only the intercepts and the slopes of the macro-level variables. The individual-level coefficients are presented in Table 4 and will be discussed in the next section.

Table 3.

Multilevel Regression Models for Housework Sharing (Log-Odds of the Country-Level Predictors).^a

	Women’s reports					Men’s reports
	Model 1	Model 2	Model 3	Model 4	Model 5	Model 1	Model 2	Model 3	Model 4	Model 5
Constant	1.682***(0.163)	1.679*** (0.159)	1.694*** (0.160)	1.697*** (0.160)	1.689*** (0.158)	−0.623** (0.228)	−0.594** (0.227)	−0.613** (0.227)	−0.611** (0.228)	−0.589** (0.227)
Men’s hours of employment/week	—	0.146*** (0.040)	—	—	0.106* (0.046)	—	−0.095** (0.030)	—	—	−0.080* (0.035)
Women 15-64 years old in workforce %	—	—	−0.128** (0.043)	—	−0.076 (0.048)	—	—	0.091** (0.031)	—	0.076* (0.037)
Gender Empowerment Measure	—	—	—	−0.130** (0.044)	−0.011 (0.056)	—	—		0.074* (0.034)	−0.030 (0.044)
Variance components
Intercept	0.040	0.018	0.022	0.025	0.012	0.011	0.004	0.004	0.007	0.001
Log-likelihood	−812.814	−807.505	−809.138	−809.040	−805.715	−660.036	−655.741	−656.357	−657.845	−653.727
Deviance	1625.628	1615.011	1618.276	1618.080	1611.431	1320.072	1311.482	1312.714	1315.690	1307.454
AIC	1669.628	1661.011	1664.276	1664.080	1661.431	1364.072	1357.482	1358.714	1361.690	1357.454
N	8,778					7,596
Groups	23

Note. AIC = Akaike information criterion. Dependent variable: proportion of domestic chores performed by each partner (unstandardized regression coefficients, standard errors in parentheses). Individual-level controls: age, marital status, number of household members, household’s hours of housework per week, presence of children, employment status, hours of employment per week, partner’s employment status, partner’s hours of employment per week, proportion of household income, relative education, value orientation.

p < .05. **p < .01. ***p < .001 (two-tailed).

Table 4.

Multilevel Regression Models for Housework Sharing (Log-Odds of the Individual-Level Predictors).

	Model 5
	Women’s Reports	Men’s Reports
Constant	1.689*** (0.158)	−0.589** (0.227)
Employed vs. not employed	0.107 (0.119)	0.054 (0.137)
Hours of employment per week	−0.012*** (0.003)	−0.012*** (0.003)
Partner employed vs. not employed	−0.234 (0.133)	−0.028 (0.119)
Partner hours of employment per week	0.015*** (0.003)	0.012*** (0.003)
Presence of children	0.142 (0.079)	−0.036 (0.084)
Proportion of income, ref: none
Very small	−0.275* (0.109)	−0.126 (0.223)
Under a half	−0.513*** (0.099)	−0.159 (0.194)
About half	−0.568*** (0.105)	−0.202 (0.188)
Over a half	−0.679*** (0.123)	−0.287 (0.188)
Very large	−0.590*** (0.163)	−0.397* (−0.196)
All	−0.435* (0.188)	−0.568** (0.200)
Relative education, ref: both low
Both medium	−0.209* (0.086)	0.141 (0.087)
Both high	−0.354*** (0.093)	0.342*** (0.095)
Partner higher	−0.154 (0.093)	0.195* (0.096)
Partner lower	−0.202* (0.092)	0.141 (0.094)
Value orientation	−0.050* (0.022)	0.073** (0.024)
Variance components
Intercept	0.012	0.001
Log-likelihood	−805.715	−653.727
Deviance	1611.431	1307.454
AIC	1661.431	1357.454
N	8,778	7,596
Groups	23

Note. AIC = Akaike information criterion. Dependent variable: proportion of domestic chores performed by each partner (unstandardized regression coefficients, standard errors in parentheses). The models include the three macro-level measures and also control for age, marital status, and the couple’s total numbers of housework as reported by the respondent.

p < .05. **p < .01. ***p < .001 (two-tailed).

Looking at Models 2, 3, and 4, the results indicate that women report spending more relative time on housework (estimate = 0.146, p < .001) in countries where men on average work longer hours. Women report spending less relative time on housework, however, if the percentage of women in the work force is high (estimate = −0.128, p < .01) and if women are highly empowered (estimate = −0.130, p < .01). Men’s reported relative time on housework, to the contrary, is lower in countries with longer standard working hours for men (estimate = −0.095, p < .01), and higher in countries with high female labor force participation (estimate = 0.091, p < .01), and high levels of gender empowerment (0.074, p < .05). Thus, the results for both women’s and men’s reports are in line with Hypotheses 1 and 2: In countries where men work more hours in the public sphere, women report spending more time on housework; in contexts where women are more present in the public sphere, they report spending less time on housework. Likewise, men report doing relatively less housework in countries where men work longer hours and more in countries where women are more empowered and more present in the labor market. Furthermore, the coefficients for female labor force participation and for gender empowerment are larger and show greater statistical power in women’s reports compared with men’s. In line with Becker’s (1981) idea of rational allocation of time, women are more responsive than men to women’s aggregate behavior. However, looking at the magnitude of the coefficient for men’s working hours, women also seem to be more receptive than men to men’s aggregate behavior.

Model 5 includes all three macro-level variables and shows that, for women, Hypotheses 3a is confirmed: the percentage of women in the work force and the GEM lose statistical significance once the presence of men in the public sphere is controlled. Indeed, the estimate for men’s standard working hours remains large and significant. This indicates that men’s working hours are quite relevant in explaining the relative time women spend on housework. By contrast, in the case of men’s reports, men’s working hours and female labor force participation are both significant and are very similar in magnitude, while only the GEM loses statistical significance. Figure 1 shows the predicted values¹² of the dependent variable, calculated from Model 5, for women and men with 95% confidence intervals at different points of the macro variables. Graph (a) in Figure 1 shows the predicted values for women and men calculated at different points of men’s average working hours. As can be seen, there is a significant association for both genders, but the slope is somewhat steeper for women, indicating that women’s reports are slightly more responsive than men’s to this macro-level characteristic. Graph (b) shows that there is a statistically significant association between female labor force participation and the relative time spent on chores for men only, while Graph (c) shows that the slope for the GEM is flat for both men and women. In other words, in the case of women, once men’s average hours of employment are controlled, measures of women’s presence in the public sphere are not associated with the allocation of time to domestic chores. By contrast, in the case of men, average working hours predict a more unequal allocation of chores, while female labor force participation is associated with more equal housework sharing.

Figure 1.

Predicted values with 95% confidence intervals of women and men’s share of housework at different values of the macro-level variables. In Graph (a), the predictions are calculated and plotted at different points of men’s weekly hours of paid work; in Graph (b), the predictions are calculated and plotted at different levels of female labor force participation; in graph (c), the predictions are calculated and plotted at different levels of gender empowerment. Values are obtained from the estimates of Model 5 and the predictions are adjusted by setting all individual-level variables to their grand means.

Individual-Level Predictors

As can be seen in Table 4, which reports the coefficients for the individual-level predictors of Model 5, the results are in line with previous studies, so I only briefly comment on them. Own employment hours and employment hours of the partner are significantly associated with housework sharing. For both women and men, the log-odds for each additional hour of employment are −0.012 (p < .001), while the coefficient for each additional employment hour of the partner is 0.015 (p < .001) for women and 0.012 (p < .001) for men. The presence of children younger than 18 years old is not associated with the partners’ share of housework for either women or men. As far as relative resources are concerned, for women, each increase in the proportion of household income, up to over a half of the total (estimate = −0.679, p < .001), is associated with a reduction in the proportion of housework.¹³ Men do not respond to relative earnings in a linear way either, as they perform a significantly lower share of housework only when they are responsible of the entire (estimate = −0.568, p < .01) or nearly the entire household income (estimate = −0.397, p < .05). For both women and men being part of a highly educated couple is a very strong predictor of equal sharing. The estimate for a highly educated woman living with a highly educated partner is −0.354 (p < .001), while for men the association is positive and of similar magnitude (estimate: 0.342, p < .001). Finally, as expected, individuals who display gender-equal attitudes tend to share time on housework more equally. For women, each increase in the gender-role variable is associated with a −0.050 (p < .05) decrease in the log-odds of sharing housework. Thus, women who display the least traditional value orientations perform 67% of the chores, that is, 5 percentage points less than the most traditional women. Traditional men, instead, do around 31% of the housework as opposed to men with more modern attitudes toward gender roles who perform 37% (estimate = 0.073, p < .01).

Conclusion and Discussion

This article investigates the relation between the allocation of time to domestic chores within couples and the presence of men in the public sphere. The article provides three main contributions to the literature. First, it includes a macro-level variable regarding the extensive presence of men in the public sphere to investigate the relative time partners spend on housework. Second, it shows that two macro-level measures used in previous research lose most, albeit not all, of their statistical power once men’s weekly average working hours are controlled. Third, it suggests that the results at both levels are not perfectly gender symmetric. This indicates that measures based on reported time, while displaying a very high degree of approximation, are not exempt from a form of gender-driven report bias (Marini & Shelton, 1993; Niemi, 1993).

People act on their personal beliefs and preferences, attitudes, and values. However, individuals can also adapt to new practices that they have identified as appropriate, either via contagion or via structural equivalence or both (Burt, 1987). Research on the division of chores within couples has shown that women and men who are highly educated or hold gender-equal attitudes share chores more equally (Davis & Greenstein, 2004; Lavee & Katz, 2002), while women do a higher share of chores if they are not employed or work few hours (Geist, 2005). Research has also brought some evidence supporting the notion that structural characteristics regarding the aggregate behavior of women and the diffusion of gender egalitarian norms of behavior are associated with a more egalitarian division of time on chores within couples (Batalova & Cohen, 2002; Fuwa, 2004; Treas & Tai, 2012).

This article, instead, provides empirical evidence of the importance of men’s aggregate behavior in explaining partners’ relative time on housework. In fact, the presence of women in the workforce, women’s degree of empowerment, and men’s working hours are all associated with the division of time on housework. However, when men’s standard working hours are controlled for, the aggregate behavior of women loses statistical power in predicting women’s reported relative time on housework. This is of particular relevance considering that previous articles found a positive association between women’s presence in the public sphere and a more egalitarian division of labor at home (Fuwa, 2004; Ruppanner, 2010). By contrast, men’s reported share of domestic chores is associated with both men’s standard working hours and female labor force participation; furthermore, the associations have just about the same magnitude.

Why would men’s working hours be the strongest Level 2 predictor of women’s reported share of housework? Given the long-standing tradition of the gendered division of labor, it could be that men’s long standard working hours are more effective in maintaining the status quo than women’s presence in the public sphere is in spurring change. In other words, women’s growing presence in the work force or women’s greater empowerment are still too weak, in comparison with men’s established presence, to move the wheel in the direction of a more egalitarian division of chores within couples. To put it differently, the contextual characteristics stemming from women’s behavior are not yet diffuse enough to justify the individual risk of adopting a behavior that deviates from standard gender norms. On the contrary, men’s long-standing and ongoing presence in the public sphere could be keeping women from adopting innovative behavior, thus stalling the equalization process. Additionally, one must consider that long standard working hours for men may not just be indirectly signaling that a gendered division of labor is the “proper” behavior: Indeed, the time binds generated by demanding workplaces could yield an aggregate absence of men at home that, regardless of male-breadwinning ideologies and of the presence of women in the public sphere, might make it difficult for men to share housework more evenly with their partners (Gerson, 2010).

Thus, in the process of the incomplete revolution (Esping-Andersen, 2009), women’s presence in the public sphere is a weak predictor of an equal division of chores in the household, indicating that Arlie Hochschild’s (1989) second shift, that is, the extra amount of unpaid labor that employed women take on when they return home from the workplace, is still in place. Furthermore, despite the diffusion of egalitarian values within the framework of the Second Demographic Transition (Lesthaeghe, 2010), men still have difficulties in putting more effort in family life because of structural constraints, while so-called “gender flexibility” still appears out of reach (Gerson 2010). Interestingly, while women’s reports are more influenced by men’s, rather than women’s, aggregate behavior, men seem to be more affected than women by women’s aggregate behavior. In fact, women’s presence in the labor market is associated with men’s reported share of housework to roughly the same extent as men’s working hours, although with opposite signs. In other words, there does seem to be an equalization process under way, but in the case of men, this is being counterbalanced by the long-standing presence of men in the public sphere (or absence from the private sphere), resulting in a still unequal allocation of time to domestic chores in the household.

Last, although I find a large symmetry in the accounts of housework by women and men, throughout the analysis, an important gender difference emerges. After controlling for individual characteristics the cross-country differences are much larger for women than for men. This points to the fact that contextual features shape women’s more than men’s perception of their share of housework. As far as domestic chores are concerned, therefore, women seem to act on prevailing norms of behavior to a greater extent than men. A tentative interpretation of this finding could lie in the fact that the changes in women’s behavior over the past decades have been much larger than men’s. Of course, men today differ from their fathers (Bianchi, Robinson, & Milkie, 2006; Gregory & Milner, 2011), but the notion that sees men as providers rather than caregivers has gone substantially unchallenged (O’Brien, Brandth, & Kvande, 2007). By contrast, the changes in women’s roles, both in the public and in the private sphere, have been massive, albeit cross-nationally very diverse (Esping-Andersen, 2009). Thus, in a time of profound and rapid change, women might be more in need than men to look for external references to seek their “appropriate behavior.” Therefore, they act on the contextual prevailing norms of behavior—even men’s aggregate behavior—more than their male counterparts. The article therefore confirms previous findings on the allocation of time to chores within couples; however, the different results that emerged by gender in the cases outlined above suggest that data collected through the response of only one member of the partnership are not a perfect measure of the actual allocation of time to chores. Notwithstanding, in the majority of cases, the results for women and men are highly symmetric. This indicates that the measure is a very good approximation of the true allocation of time to domestic chores within couples, although we can only refer to it as the perception of relative time allocated to chores. Without comparable, high-quality, household-level time-budget data on a large number of countries, this is the closest approximation researchers can get of the true allocation of time to chores within couples.

Finally, above and beyond the expected finding that the division of chores within European couples in 2010 is still strongly gendered, the article shows that men’s aggregate behavior is a key element in the allocation of time to domestic chores in European countries. In the light of this, including men’s aggregate behavior in the analysis of household labor may prove fruitful for future research.

Footnotes

Acknowledgements

The author would like to thank the anonymous reviewers for their useful comments and suggestions and Judith Treas for her valuable remarks on an early version of this article.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was partially funded by FP7 ERC-2010-StG: Families of Inequalities—Social and Economic Consequences of the Changing Work–Family Equilibria in European Societies.

Notes

References

Batalova

Cohen

P. N.

(2002). Premarital cohabitation and housework: Couples in cross-national perspective. Journal of Marriage and Family, 64, 743-755.

Becker

G. S.

(1981). A treatise on the family. Cambridge, MA: Harvard University Press.

Bianchi

Robinson

Milkie

(2006). Changing rhythms of American family life. New York, NY: Russell Sage Foundation.

Blair

S. L.

Lichter

D. T.

(1991). Measuring the division of household labor: Gender segregation of housework among American couples. Journal of Family Issues, 12, 91-113.

Blood

Wolfe

(1960). Husbands and wives: The dynamics of married living. New York, NY: Free Press.

Brayfield

(1992). Employment resources and housework in Canada. Journal of Marriage and the Family, 54, 19-30.

Breen

Cooke

L. P.

(2005). The persistence of the gendered division of domestic labour. European Sociological Review, 21, 43-57.

Brines

(1994). Economic dependency, gender, and the division of labor at home. American Journal of Sociology, 100, 652-688.

Bühlmann

Elcheroth

Tettamanti

(2010). The division of labour among European couples: The effects of life course and welfare policy on value–practice configurations. European Sociological Review, 26, 49-66.

10.

Burt

R. S.

(1987). Social contagion and innovation: Cohesion versus structural equivalence. American Journal of Sociology, 92, 1287-1335.

11.

Coleman

(1991). The division of household labor: Suggestions for further empirical consideration and theoretical development. In Blumberg

R. L.

(Eds.), Gender, family and economy: The triple overlap (pp. 245-260), Newbury Park, CA: Sage.

12.

Connell

(1985). Theorizing gender. Sociology, 19, 260-275.

13.

Craig

Mullan

(2010). Parenthood, gender and work-family time in the United States, Australia, Italy, France, and Denmark. Journal of Marriage and Family, 72, 1344-1361.

14.

Davis

S. N.

Greenstein

T. N.

(2004). Cross-national variations in the division of household labor. Journal of Marriage and Family, 66, 1260-1271.

15.

Dribe

Stanfors

(2011). Does parenthood strengthen a traditional household division of labor? Evidence from Sweden. Journal of Marriage and Family, 71, 33-45.

16.

Eeckhaut

M. C. W.

Van

Putte

Gerris

J. R. M.

Vermulst

A. A.

(2013) Analysing the effect of educational differences between partners: A methodological/theoretical comparison. European Sociological Review, 29, 60-73.

17.

Esping-Andersen

(2009). The incomplete revolution: Adapting to women’s new roles. Cambridge, England: Polity Press.

18.

European Social Survey. (2010). Round 5 data (Data file edition 2.0). Bergen, Norway: Norwegian Social Science Data Services (Data archive and distributor of ESS data).

19.

Eurostat. (2012). Employment rate, by sex. Retrieved from http://epp.eurostat.ec.europa.eu/tgm/table.do?tab=table&plugin=1&language=en&pcode=tsdec420

20.

Evertsson

Nermo

(2004). Dependence within families and the division of labor: Comparing Sweden and the United States. Journal of Marriage and Family, 66, 1272-1286.

21.

Ferree

M. M.

(1990). Beyond separate spheres: Feminism and family research. Journal of Marriage and Family, 52, 866-884.

22.

Fuwa

(2004). Macro-level gender inequality and the division of household labor in 22 countries. American Sociological Review, 69, 751-767.

23.

Fuwa

Cohen

P. N.

(2007). Housework and social policy. Social Science Research, 36, 512-530.

24.

Geist

(2005). The welfare state and the home: Regime differences in the domestic division of labour. European Sociological Review, 21, 23-41.

25.

Gershuny

(2000). Changing times: Work and leisure in postindustrial society. Oxford, England: Oxford University Press.

26.

Gerson

(2010). The unfinished revolution. How a new generation is reshaping family, work and gender in America. New York, NY: Oxford University Press.

27.

Greenstein

T. N.

(2000). Economic dependence, gender, and the division of labor in the home: A replication and extension. Journal of Marriage and the Family, 62, 322-335.

28.

Gregory

Milner

(2011). What is “new” about fatherhood?: The social construction of fatherhood in France and the UK. Men and Masculinities, 14, 588-606.

29.

Hiller

(1984). Power dependence and division of family work. Sex Roles, 10(11-12), 1003-1019.

30.

Hochschild

(1989). The second shift: Working parents and the revolution at home. New York, NY: Viking.

31.

Hook

J. L.

(2006). Care in context: Men’s unpaid work in 20 countries, 1965-2003. American Sociological Review, 71, 639-660.

32.

Hook

J. L.

(2010). Gender inequality in the welfare state: Sex segregation in housework, 1965-2003. American Journal of Sociology, 115, 1480-1523.

33.

Hox

(2010). Multilevel analysis, techniques and applications. New York, NY: Routledge.

34.

Huber

Spitze

(1983). Sex stratification: Children, housework and jobs. New York, NY: Academic Press.

35.

International Labour Organization. (2010). LABORSTA Internet. Retrieved from http://laborsta.ilo.org

36.

Juster

Stafford

(1991). The allocation of time: Empiric findings, behavioral models, and problems of measurement. Journal of Economic Literature, 29, 471-522.

37.

Kerman

Gelman

(2007). Manipulating and summarizing posterior simulation using random variable objects. Statistics and Computing, 17, 235-244.

38.

Knudsen

Wærness

(2008). National context and spouses’ housework in 34 countries. European Sociological Review, 24, 97-113.

39.

Kühhirt

(2012). Childbirth and the long-term division of labour within couples: how do substitution, bargaining power, and norms affect parents’ time allocation in West Germany? European Sociological Review, 28, 565-582.

40.

Lavee

Katz

(2002). Division of labor, perceived fairness and marital quality: The effect of gender ideology. Journal of Marriage and Family, 64, 27-39.

41.

Lesthaeghe

(2010). The unfolding story of the second demographic transition. Population and Development Review, 36, 211-251.

42.

Lewin-Epstein

Stier

Braun

(2006). The division of household labor in Germany and Israel. Journal of Marriage and Family, 68, 1147-1164.

43.

Marini

Shelton

(1993). Measuring household work: Recent experience in the United States. Social Science Research, 22, 361-382.

44.

McDowell

Cox

(2004). How does one do regression when the dependent variable is a proportion? Retrieved from http://www.ats.ucla.edu/stat/stata/faq/proportion.htm

45.

Nazio

Blossfeld

H. P.

(2003). The diffusion of cohabitation among young women in West Germany, East Germany and Italy. European Journal of Population, 19, 47-82.

46.

Niemi

(1993). Systematic error in behavioral measurement: Comparing results from interview and time budget studies. Social Indicators Research, 30, 229-244.

47.

O’Brien

Brandth

Kvande

(2007). Fathers, work and family life. Community, Work & Family, 10, 375-386.

48.

Presser

H. B.

(1994). Employment schedules among dual-earner spouses and the division of household labor by gender. American Sociological Review, 59, 348-364.

49.

Röhler

K. A.

Huinink

(2010). Pair relationships and housework. In Treas

Drobnič

(Eds.), Dividing the domestic (pp. 192-213). Stanford, CA: Stanford University Press.

50.

Ruppanner

(2010). Cross-national reports of housework: An investigation of the gender egalitarianism measure. Social Science Research, 39, 963-975.

51.

Ruppanner

(2012). Housework conflict and divorce: A multi-level analysis. Work, Employment and Society, 26, 638-656.

52.

Sayer

(2010). Trends in housework. In Treas

Drobnič

(Eds.), Dividing the domestic (pp. 19-38). Stanford, CA: Stanford University Press.

53.

Schober

P. S.

(2013). The parenthood effect on gender inequality: Explaining the change in paid and domestic work when British couples become parents. European Sociological Review, 29, 74-85.

54.

South

S. J.

Spitze

(1994). Housework in marital and nonmarital households. American Sociological Review, 59, 327-347.

55.

Treas

Drobnič

(Eds.). (2010). Dividing the domestic. Stanford, CA: Stanford University Press.

56.

Treas

Tai

(2012). How couples manage the household: Work and power in cross-national perspective. Journal of Family Issues, 33, 1088-1116.

57.

United Nations Development Programme. (2007). Human development report 2007/2008. New York, NY: Palgrave Macmillan.

58.

van der Lippe

de Ruijter

Raub

(2011). Persistent inequalities in time use between men and women: A detailed look at the influence of economic circumstances, policies, and culture. European Sociological Review, 27, 164-179.

59.

Voicu

Strapcova

(2009). Housework and gender inequality in European countries. European Sociological Review, 25, 365-377.

60.

West

Zimmerman

D. H.

(1987). Doing gender. Gender & Society, 1, 125-151.

61.

Yodanis

(2005). Divorce culture and marital gender equality: A cross-national study. Gender & Society, 19, 644-659.