Is Cohabitation More Egalitarian? The Division of Household Labor in Five European Countries

Abstract

This article compares the gendered allocation of household labor between married and cohabiting couples in five European countries: Germany, France, the United Kingdom, Italy, and Spain, testing whether cohabitors show more egalitarian divisions of labor and hypothesizing that the effect of cohabitation differs across countries, depending on the baseline equality and on the meaning of cohabitation. In order to examine to what extent there is equality, not only each partners’ contribution to the total housework time is considered but also who does what: Some tasks are more constraining than others, and gender and partnership differences specific to those tasks are investigated too. The empirical analysis is based on Multinational Time Use Surveys (N = 58,490), using ordinary least squares linear regression models. Results show that cohabiting couples have a more egalitarian division of labor but that there are important country differences.

Keywords

cohabitation domestic labor gender roles marriage time use

Introduction

Reproductive work, defined as “unpaid work done to maintain family members and/or a home” (Coltrane, 2000), is key for the good functioning of society, and because of its special characteristics, it has often been presented as opposed to productive or paid work. Historically, housework has constituted a gendered activity, being ascribed to the private sphere and considered women’s specialty. In the late 20th century, the generalized increase in female labor force participation lead some authors to predict the end of this gendered specialization, but in spite of cross-country evidence pointing at a higher involvement of men in housework, an egalitarian distribution seems hard to achieve (Davis & Greenstein, 2004).

The persistent gendered division of labor has induced sociologists to question the reasons of such division as well as to investigate under what conditions more egalitarian distributions of housework emerge. Studies on the division of household labor have focused especially on married couples, given that heterosexual marriage was the normative form of partnership until very recently, but with the spread of alternative living arrangements, the unit of analysis has widened in order to include same sex and cohabiting couples. Specifically, previous research on cohabitation has shown that nonmarried couples have a more egalitarian division of labor than marrieds (e.g., Batalova & Cohen, 2002; Coltrane, 2000).

Most of the literature on cohabitation and housework has treated this effect of cohabitation as an empirical regularity that works similarly across countries. However, it is well known that cohabiting relationships may have different meanings, which vary significantly across countries (Heuveline & Timberlake, 2004). In this sense, this article contributes to the existing literature in two ways: first, by assessing the effects of cohabitation in different countries using harmonized time use data, and second, by taking into account that its effect may not be homogeneous but related to the role that cohabitation plays in each society. The second aim of this research is more exploratory and will try to derive testable hypothesis from our theoretical and empirical knowledge about the division of labor and living arrangements.

The article is structured as follows: First, the literature on different partnerships and the division of labor is reviewed, deriving testable hypotheses. The second section describes the data (Time Use Surveys) and discusses the methods available for the statistical analysis, whose results are presented in the third section. Given the focus on country differences, a significant space of the empirical section is devoted to describing cross-country variations in the distribution of the main variables under study.

Domestic Work, Gender Roles, and Living Arrangements

In the 1990s, the study of domestic work became an important area of social science studies, and the relevance of its relationship to other social phenomena, such as family relations, labor markets, values, and the life cycle was soon established (Coltrane, 2000). This type of unpaid work presented a strong gender asymmetry, along the model of the male breadwinner, according to which the husband worked outside the home and the wife was responsible for the home and children. Such specialization pattern was dominant during most of the 20th century, and several sociological theories offered explanations for its persistence. Even though there are important variations, these explanations can be classified into two groups: theories that focus on relative resources and theories that point at socialization processes.

Relative resources theories date back to Becker’s (1981) New Household Economics, who interpreted women’s specialization in household tasks as a result of a rational calculus made by the household as a unit and resulting from men’s comparative advantage on human capital and labor market characteristics. This portrait of the division of labor inside the family was refined through the ideas of negotiation and bargaining (Lundberg & Pollak, 1996), where household decisions were not made according to a unitary utility function, because individual members have their own preferences and must bargain with each other to reach equilibrium. Individual bargaining power depends on the potential contribution to the household, which is usually measured in terms of earnings. If women’s bargaining power was higher than men’s, alternative specialization models could emerge, and therefore these theories are considered gender neutral. Empirical evidence has indeed shown that women in full-time paid work contribute less to housework than women who do not work or work fewer hours; and that double income couples show a more egalitarian distribution of tasks (Gershuny, 2000). Results have been less clear for men, although those who work fewer hours have also been found to do more housework (Bianchi, Milkie, Sayer, & Robinson, 2000). However, in some cases full-time employed women increase their contribution to domestic chores when their partner loses the job (Brines, 1994), and this is difficult to explain from the relative resources perspective.

The second group of theories, and an alternative explanation to relative resources, focuses on socialization processes and gender role formation. According to these theories, women’s higher involvement in domestic work is not due to the characteristics of the labor market or to a rational calculus made by the household, but to their own gender identity and existing social norms. Gender identity is formed from childhood on, along a complex process in which many institutions play a role: through this process, women continuously see and learn that higher domestic skills are expected from them. Developing this point, doing gender theories argue that gender identity is built through everyday interactions (West & Fenstermaker, 1993), and women continue to do more housework because it is associated to feminity, as part of the confirmation of their own identity. Gender theories have also found empirical support for their claims: the literature has shown that couples with egalitarian gender values share their housework in a more egalitarian fashion (Lennon & Rosenfield, 1994; Lück & Hofäcker, 2003; Meil Landwerlin, 2005), and that highly educated women contribute less to housework, whereas the inverse effect is found for men: more education implies more housework on their side (Anxo, 2002; Bianchi et al., 2000; Gershuny, 2000; Perkins & de Meis, 1996; Pittman, 1995). Despite this, evidence related to education must be interpreted cautiously, because it may either be an indicator of egalitarian values or a proxy for social class, human capital, and bargaining power—when homogamy levels are taken into account.

In addition to ascertaining the importance of socioeconomic variables and gender values on domestic work performance, comparative studies have also found national differences in the division of labor. After controlling for the aforementioned factors, some countries show more egalitarian divisions of domestic labor than others, and the reasons for such differences are not clearly established. The possible influence of the welfare states typology has received only partial confirmation: Nordic states show more egalitarian divisions of domestic work but conservative states are very heterogeneous (Bühlmann, Elcheroth, & Tettamani, 2010; Geist, 2005). Gershuny and Sullivan (2003) conclude that welfare states and policies clearly determine participation in paid work but that unpaid work is related to broader determinants that research has yet to pinpoint. Other macrolevel variables, such as aggregate levels of women’s labor force participation (Hook, 2006), economic development, and gender norms (Fuwa, 2004), have also shown a significant effect in multilevel models that try to explain this cross-country variation. These studies have pointed out the importance of considering the role of the macro level as well as couples’ or individual characteristics to explain the gendered allocation of housework.

But countries not only differ on their average share of domestic work by sex, they also differ on the popularity of alternative forms of partnership. With the diffusion of new family models, the type of partnership became a variable of interest for studies on the division of domestic work and results have consistently shown that cohabiting couples hold a more egalitarian division of housework than marrieds (Batalova & Cohen 2002; Baxter, 2005; Shelton & John, 1996; South & Spitze, 1994). Even though this empirical regularity is well established, the reasons for it are not so obvious. On the one hand, cohabitation may be selective of couples with special traits, which make them less prone to specialization. For instance, consensual unions have been found to hold more egalitarian gender values than marrieds (Clarkberg, Stolzberg, & Waite, 1995), which may definitely lead them to more egalitarian housework allocations, although longitudinal studies have found a trend toward a more traditional division of labor when cohabitors marry or have children (Gupta, 1999). Cohabitation also selects couples with different socioeconomic profiles than marriage (Kiernan, 2002), and given that socioeconomic variables affect significantly the division of labor, as mentioned above, consensual unions may distribute domestic chores more equally as a consequence of these divergent profiles—and not necessarily of being more progressive. Cohabitation also presents lower exit costs than marriage (Sigle-Rushton & McLanahan, 2002), which may lead to more couples with lower commitment levels cohabiting instead of marrying, as well as to lower incentives for specialization. On the other hand, something may happen inside the relationship that fosters a different division of labor: For instance, the duration of cohabiting unions, usually shorter than that of marriages, may reduce women’s interest in specializing in reproductive work (Clarkberg, 1999). Unmarried co-residence may also be used as part of partners’ selection process, where willingness to share housework may be considered a desirable trait (Breen & Cooke, 2005), and especially so in countries with a high degree of gender inequality (Ono, 2003). Cohabitation may even be the result of a negotiation and reflect an asymmetry in bargaining power, which again would influence the domestic division of labor.¹

The latter explanation is related to the role of cohabiting unions and the meaning that individuals attribute to them in their partnership biographies. According to the classification presented by Heuveline and Timberlake (2004), cohabitation may play six roles, and in each country one of those roles prevails: in Southern Europe consensual unions would be defined as marginal (not prevalent and publicly discouraged); in Central Europe they would act as a stage in the marriage process or as a previous stage to marriage (cohabitation being a childless stage often ending in marriage)²; in contrast, in France they would be an alternative to marriage (with a high incidence of long-term cohabitation and childbearing). In Anglo-Saxon countries, it can be considered an alternative to singlehood (relatively high incidence, but with few long-term unions and transitions into marriage).³ An additional possibility is that cohabitation is indistinguishable from marriage, as it happens in Scandinavian countries.

Taking into account these different meanings, is it reasonable to expect cohabitation to have the same effects on the division of domestic work for all countries? The literature has not considered this issue in a systematic manner, and therefore only speculations based on previous research results can be made here, but we can try to systematize the main factors (or groups of factors) identified by the literature as plausible mechanisms rendering cohabitors more egalitarian by distinguishing four factors: socioeconomic profiles (e.g., highly educated couples are more egalitarian), life-course stages (e.g., there seems to be some traditionalization after childbirth), values, and mate selection preferences. The first two are easily measured in most quantitative data sets but the last two are often difficult to account for. In addition to this, we should bear in mind the degree of gender equality in each country because the equalizing effect of cohabitation or any other variable will probably be smaller in countries where domestic tasks are already shared more equally. We can then go back to Heuveline and Timberlake’s typology and check whether the aforementioned factors are similarly distributed among cohabitors across countries.

To begin with, we may expect no effect of cohabitation in Scandinavian countries, where cohabitation is “indistinguishable” from marriage, given the similarities between both living arrangements and the high level of gender equality. In contrast, in Southern European countries, cohabitation is “marginal” and highly selective of women with high educational attainment and active in the labor market, who choose to deviate from the traditional process of family formation, and consensual unions are mainly premarital (Meil Landwerlin, 2003). The gender context in these countries is quite conservative, and it has been hypothesized that cohabitation may help women select more egalitarian partners. Thus, all the aforementioned factors would point at cohabitation being a more egalitarian living arrangement.

In countries where cohabitation is a “stage in the marriage process,” the life stage of consensual unions is quite clear: most couples are childless, relatively young, and will end up marrying (Heuveline & Timberlake, 2004). There is no clear pattern of selection into consensual unions according to socioeconomic traits. Overall, gender equality in these countries is higher than in Southern Europe but lower than in Northern Europe. Regarding mate selection, cohabitation could produce egalitarian effects if women valued willingness to share housework in their potential partners. However, the opposite effect could be found if women’s domestic skills were a highly valued trait instead. Concerning attitudes, cohabitation may select individuals with less traditional values, which would have an egalitarian effect. Thus, we would expect cohabitation to have a limited influence, related to attitudes and life stage.

The case of cohabitation as an “alternative to singlehood”—it could also be considered a lesser bond (Schoen & Weinick, 1993)—is problematic. Partners may not feel much pressure to be considered potential spouses and men may find no incentive to share tasks, but at the same time, women may find no incentive in specializing. Previous research on the U.S. case has shown that cohabiting women spend less time doing household chores than marrieds, although men’s performance did not differ much by living arrangement (South & Spitze, 1994). In this case, there is no clear association between the identified factors and our empirical knowledge about this type of consensual unions, which leads us to expect no significant effect of cohabitation on the division of housework.

Cohabiting unions can also act as an “alternative to marriage.” In those cases, partner selection has already been accomplished, and we will find cohabitors in similar stages of life than marrieds. This case is similar to cohabitation in Scandinavian countries, but here there is an important difference between both living arrangements in terms of the legal and possibly social consequences of both partnerships. They are not indistinguishable but, in spite of this, some couples decide to raise their family outside marriage. The reasons for doing so may be varied, but in the French case, there seems to be a desire to depart from traditional families and roles (Leridon & Toulemon, 1995), and thus cohabiting unions may also have more egalitarian arrangements than marrieds, controlling for other factors. The above mentioned mechanisms and the expected effect on the division of domestic chores are summarized in Table 1.

Table 1.

Summary of the Factors Affecting the Division of Household Labor and Expected Effects for Different Types of Cohabitation.

Types	Socioeconomic Profile	Life Stage	Mate Selection	Attitudes	Effects on Equal Division
Indistinguishable	=	=	=	=
Marginal	More educated	Childless	Yes	Less traditional	++
Alternative to marriage	=	=	=	Less traditional	+
Previous stage	=	Childless	Yes	Slightly less traditional	+
Alternative to singlehood	=	Childless, young people	Not necessarily	Less traditional	?

Up to this point, housework has been considered as a general concept, and this may not be the most appropriate strategy because domestic chores involve many different activities. In this article, the focus will be on household tasks and we will not consider childcare. The idea behind this selection is that both types of work are considered differently by family members: Household tasks are often routine tasks, performed out of necessity, many times deemed boring. Childcare tasks, however, have also an emotional component attached to them, and the literature has shown that indeed parents do not try to bargain with child care, but rather enjoy activities together with their children, thus considering child care closer to leisure in some respects (Hallberg & Klevmarken, 2003; Sayer, Bianchi, & Robinson, 2004).

But even if the analysis is restricted to household tasks, this concept includes different chores with varied characteristics: cooking, ironing, or shopping for groceries are not equally routine or time-constraining activities. Cleaning tasks (cleaning the house, washing dishes, or doing the laundry) are considered central, routine, and repetitive tasks, as opposed to gardening or shopping, which are performed occasionally, are closer to leisure, and can be more time flexible and easy to postpone. The gendered division of household tasks also follows these lines, and it has been observed that, on average, women concentrate more on routine tasks and men focus on more leisurely tasks; to such an extent that some authors use the label “female tasks” to denote the former and “male tasks” to refer to the latter (Bianchi et al., 2000; Thompson & Walker, 1989). Therefore, if the division of tasks is to be investigated, it will be more informative to consider who does what in the couple and to what extent, instead of only measuring how much individuals do.

To sum up, and taking into account previous theory and findings, the hypotheses to be tested in the empirical section of this article are the following:

Hypothesis 1: Women will contribute to housework more in all unions but unmarried couples will have a more egalitarian distribution. In two senses:

Hypothesis 1a: Cohabiting women will contribute to housework less than marrieds whereas cohabiting men will contribute more than married men.

Hypothesis 1b: Men in consensual unions will do higher proportions of female tasks than marrieds.

Hypothesis 2: Because of the different levels of both gender equality in the private sphere and natures of the relationship, cohabitation effects will differ by country. The specific effect will depend on the configuration of socioeconomic status, attitudes, life stage, and mate selection processes that cohabitation entails in each country, as summarized in Table 1.

Method

Data and Case Selection

The data used in this research were extracted from the Multinational Time Use Surveys (MTUS; Gershuny et al., 2010) Project, which harmonizes time use surveys carried out by national statistics institutes in all participating countries. In addition to providing information on background socioeconomic variables and household characteristics, respondents of these surveys are asked to fill in a diary of activities. The diary divides the day into 10- or 15-minute periods, and respondents have to inform on what activity they were performing during each period. Only some of these surveys collect data on both members of the couple, and this has determined the sample of European countries available for this study: Spain, Italy, Germany, France, and the United Kingdom. Fortunately enough, these countries represent the most interesting meanings of cohabitation, with the exception of Scandinavian countries, for which no data were available at the MTUS. Data collection took place in 2002-2003 for Spain and Italy (N = 46,774 and N = 51,206, respectively), in 2001-2002 for Germany (N = 11,949), 2000-2001 for the United Kingdom (N = 15,467), and 1998-1999 for France (N = 15,441). After selecting both couple members who were either married or cohabiting in all countries, the total sample adds up to 58,490 individuals.

Method

Concerning regression techniques, the use of a multilevel model was discarded in this article, mainly because of the small number of countries in the sample, which does not allow for enough variation on country-level variables. The main aim of this research is to assess the impact of cohabitation on the individual’s (Level 1) contribution to domestic work in different countries (Level 2). A multilevel analysis would determine to what extent the variance is explained by the different levels, as well as whether the effect of cohabitation differs by country, but given the absence of country-level variables, it would not add substantial meaning to a linear regression analysis, because it would not identify specific country effects, which is one of the points of interest in this research.

When analyzing time use data, a common problem is that many individuals do not spend any time at all doing the activity that the researcher is interested in. As a result, the dependent variable is not normally distributed because of a large number of observations with the value zero, and regression analysis is problematic. To account for these observations, Tobit regression is commonly used in time use studies, but its comparative advantages versus ordinary least squares (OLS) are also subject to discussion. In this article, OLS is preferred to Tobit because there is no theoretical reason that leads to predict some type of selection into the value 0, and because the large random sample makes OLS results less biased (Stewart, 2009). The analyses were run using Tobit too (results available from the author) and no significant differences were found. The estimated equation for the models is therefore

Y = α + β X_{i} + u_{i},

where the dependent variable y measures the amount of tasks that each individual performs and X_i comprises a set of independent variables. Data from the five countries are pooled in a single model, stepwise. In the first step, the model controls for cohabitation and country, and in a second stage, the interaction between cohabitation and country is introduced to investigate the hypothesized different effects. Men and women are analyzed separately because the independent variables are expected to produce different effects by sex.

Variables

Time use diaries provide information on how many minutes per day are devoted to specific household tasks, which provides a good approximation to the real dedication to domestic chores. The MTUS harmonizes the information and aggregates data from specific activities into broader types of domestic chores. According to the hypotheses, significant differences are expected between men and women in tasks that are more repetitive and time constraining, such as doing the laundry and cleaning, less significant differences in the tasks that are closer to leisure. Thus, three types of tasks are considered here: the first two are representatives of typically male and female chores: cleaning (includes cleaning the house, washing dishes, doing the laundry, and all related tasks—Variable AV7 in the MTUS) and occasional tasks (gardening, doing reparations, training pets—Variables AV8 and AV9 in the MTUS). The latter variable also includes occasional care for adults, which may bias some results as it is a typically female task. The third variable is a summary variable measuring the total time devoted to domestic work, including the former activities as well as shopping and cooking—Variable AV10 in the MTUS.⁴

The dependent variable could be measured in two different ways. The first one considers the number of minutes that each individual devotes to the specific chores and provides information on whether cohabitors do more or less housework than marrieds. The second one measures the percentage contribution of each individual to the household total time. This codification takes into account the actual sharing of the tasks and not simply the amount of time devoted to them, which can be high but equal to the partner’s or low and unequal. The use of two different dependent variables would add unnecessary complexity to the analysis, and in this article, we will focus on the second one: the distribution of housework. We have also analyzed the amount of time dedicated to domestic chores, and results from that analysis will be mentioned at some points in order to illustrate if more egalitarian distributions are associated to women’s lower or to men’s higher investments in household labor.

The main covariates are marital status (measured with a dummy variable with a positive outcome for cohabitation), country of residence, and the interaction between both variables. Other independent variables of interest are related to the main factors influencing the division of household tasks: educational attainment (primary, secondary, or college) and activity in the labor market (full time, part time, or not active) as well as partners’ characteristics (educational attainment and participation in the labor market) allow us to control for socioeconomic status. These variables may also reflect gender attitudes and bargaining power. In order to consider relative resources, educational homogamy between partners was included as an alternative specification for educational attainment. Results for educational homogamy are not shown here because in these models, individual educational level was more relevant than educational homogamy: college education increases men’s contribution to domestic chores and decreases women’s, whereas women’s primary education has the opposite effect, irrespective of partner’s education.

To account for the life stage of individuals, we include age (corrected by age squared), presence of children younger than 14 years, and number of children younger than 6 years. Unfortunately, we have no data on values or attitudes, and this will have to be borne in mind when interpreting results. Thus, the models control for two of the factors associated with cohabitation and influencing housework performance: life stage and socioeconomic status. If cohabitation has an effect after controlling for these factors, we would attribute it to gender values and mate selection.

The baseline model considers that both couple members are working full time, thus the dependent variable will provide insight into equity inside the relationship. For routine and female tasks, it will be considered that both decreases in women’s share or time and increases in men’s share or time lead to fairer divisions of housework; whereas for occasional tasks, the interpretation will be the opposite. For the models where the dependent variable was the percentage contribution to household tasks, an additional control is included measuring the total amount of time that the household devotes to domestic chores because the relative work load may be distributed differently depending on the size of the total load. The reference for country comparisons in all models is the United Kingdom, because in the descriptive analysis, it was shown to be the case with the most egalitarian division of housework. According to statistical tests, countries were significantly different from each other irrespective of the reference category.

Results

The five countries examined are heterogeneous in what concerns cohabitation and family formation patterns, as mentioned in the first part of this article. Such differences emerge clearly in the MTUS data. France is the country where cohabitation is more present from the cross-sectional perspective, with 20% of those in union cohabiting without marriage. The United Kingdom also shows a relatively high level of cohabitation, with a 14%, whereas in the other three countries (Germany, Italy, and Spain), cohabitation is not very extended and does not even reach a 3.5% in the Italian case. However, if only young couples are considered, the extension of cohabitation changes remarkably: the ranking of countries remains unaltered, but the proportion of cohabiting unions increases in all of them: 55% of couples cohabit in France, around 40% in the United Kingdom, 35% in Germany, 22% in Spain, and 10% in Italy.

One of the theoretical bases of this research assumes that cohabitors and marrieds behave differently, or that each partnership selects individuals who are significantly diverse. These differences are visible in Table 2, which summarizes the distribution of domestic work by sex and living arrangement, taking into consideration variables, such as age, educational attainment, labor status, and children. These data present average trends for the whole sample, the distribution by country not being shown here for the sake of simplicity. Cohabitors are on average 10 years younger than marrieds, which may not only be due to a different diffusion of cohabitation across cohorts, but also to cohabitation meaning a previous step to marriage, as already mentioned. Nonmarried couples also show higher educational levels than marrieds, salient at the college level and more significant for women, although this difference was not found among cohabitors in the United Kingdom, where cohabitation seems to be more popular among those with secondary education (data not shown). The gap in education might be related to the age structure of both living arrangements, but it is not completely explained by age composition. This profile of higher education is coherent with the higher rates of labor market participation among cohabiting women as well as with lower inactivity and more full-time enrolment, in this case also for the United Kingdom. For men, the differences in labor market involvement are small and most likely related to the age composition of each group. Finally, the last variable in this descriptive table controls for the presence of children in the household and also points at an important difference between unions, namely that childbearing is more frequently associated with marriage, even though the percentage of cohabitors with children is more than 40% and therefore not low. Italy was the case where the differences in terms of childbearing were most significant (only 34.3% of cohabitors had children).

Table 2.

Characteristics of the Sample and Division of Domestic Work by Sex and Partnership.

	Married		Cohabiting		Total
	Men	Women	Men	Women	Men	Women
Age (in years)	46.24	43.6	36.25	34.2	45.51	43.92
Education
Primary	19.36	20.95	14.29	15.73	18.99	20.57
Secondary	58.58	60.53	54.32	50.81	58.27	59.83
College	21.6	18.1	30.59	32.7	22.26	19.15
Activity
Full time	68.31	31.47	72.45	49.32	68.61	32.75
Part time	1.57	13.05	2.67	13.94	1.65	13.11
Unknown hours	11.89	8.07	10.37	6.15	11.78	7.94
Inactive	18.23	47.41	14.5	30.59	17.96	46.2
Children	54.25		43.12		51.95
Domestic work
Total time
Housework	97.62	298.54	92.84	204.59	97.27	291.78
Cleaning	17.38	118.83	18.77	77.45	17.48	115.85
Cooking	20.28	117.67	24.13	78.76	20.56	114.87
Shopping	20.18	35.52	21.42	30.42	20.28	35.15
Occasional	39.77	26.51	28.5	17.94	39.94	25.89
Proportion
Housework	23.06	76.13	29.97	70.03	23.52	76.44
Cleaning	12.93	87.16	20.35	79.65	13.43	86.57
Cooking	15.85	84.44	25.32	74.68	16.52	83.48
Shopping	32.62	67.41	37.18	62.82	32.92	67.08
Occasional	53.39	46.91	52.62	47.38	53.34	46.66

Note: All values are weighted.

The lower part of Table 2 summarizes the distribution of domestic work among couples, where “total time” presents the number of minutes devoted to each task, and “proportion” shows the percent contribution of each couple member. If the total time devoted to housework is analyzed, the difference between men and women is remarkable, but there are also important differences among unions: Cohabitors invest less time in housework than marrieds, the gap for men being quite small (around 5 minutes) but very important in the case of women (more than an hour). Men in cohabiting unions spend less time than married men doing household tasks, but when specific tasks are considered, they spend more time than marrieds in routine or female tasks and much less in occasional, male tasks. Cohabiting women consistently spend less time in all tasks than married women. The distribution of tasks between both members of the couple also points at a fairer distribution in cohabiting than in married unions, although women’s share is more than 70% for both union types. The distribution seems to follow the female/male tasks typology: women do higher proportions of cleaning and cooking whereas men do higher percentages of the chores as we move from more routine to less routine tasks.

The data in Table 2 pool all countries together; however, there are important differences concerning housework performance across countries. On the one hand, the average time per day devoted to housework by couples differs by country: Italian couples dedicate 419.8 minutes, Spanish ones 386.5, Germans 375.6, English couples 352.4, and French couples 344.1 minutes. On the other hand, there are also differences on the equality of the division. Figure 1 summarizes the distribution of housework, by sex and country and for all types of task. If all housework is considered, the United Kingdom would rate as the most egalitarian country, followed by France, Germany, and as more inegalitarian cases, Italy and Spain. The difference is significant: Men in the United Kingdom do on average one third of the housework whereas in Italy, they hardly manage a 20%. This pattern is the same for all tasks considered, except for shopping—where no pattern is found—and for occasional tasks, which are clearly male tasks: men do more than half of them in all countries. This country variation could be attributed to an array of factors that were mentioned in the first section of this article: welfare state typologies (Mediterranean countries are less egalitarian, the liberal state is the more egalitarian, and conservative countries are heterogeneous); different levels of female labor force participation; and prevailing gender norms.

Figure 1.

Percentage contribution to specific tasks, by sex, task, and country.

This descriptive evidence is favorable to the first hypotheses of the article, but descriptive results may be hiding correlations with other variables, such as women’s labor force participation. To control for socioeconomic variables and composition effects, and to examine the separate effects of cohabitation predicted by the second hypothesis, a regression analysis is needed. Table 3 summarizes results from OLS regressions with the percentage contribution to housework as dependent variable and where the independent variables of interest are partnership status (cohabitation or marriage) and country. Models were run for men and women separately because the independent variables were expected to have different effects by sex, but results are symmetrical and thus only results for women’s contribution to domestic chores are presented here. Table 3 includes two models: Model 1 controls only for cohabitation and country of residence, and Model 2 adds the other independent variables.

Table 3.

Results for Ordinary Least Squares Regression on Women’s Dedication to Housework, by Task.

	Model 1			Model 2
	Total	Cleaning	Occasional	Total	Cleaning	Occasional
Cohabitation	4.79***	6.38***	−2.89*	2.04***	3.03***	−3.95***
Country
Germany	−1.8***	−2.26	−4.32***	−0.54	−1.30*	−0.45
France	−8.26***	−11.40	5.06***	−7.52***	−11.18***	6.25***
United Kingdom	ref	ref	ref	ref	ref	ref
Italy	−14.57***	−13.50***	−4.27***	−11.95***	−10.57***	−2.83***
Spain	−11.89***	−10.72***	1.02	−1.01***	−9.11***	1.93***
Age				−0.02	−0.09	0.52
Age squared				0.00	0.00	0.00
Education
Primary				−1.60***	−1.93***	−0.91
Secondary				ref	ref	ref
College				0.99***	1.96***	−4.67***
Activity
Full time				ref	ref	ref
Part time				7.17***	8.23***	7.00***
Inactive				−3.36***	−1.38***	−9.49***
Number of children				−0.98***	−0.71***	2.91***
Children <6 years				3.67***	2.41***	3.79***
Weekend				4.49***	3.52***	9.90***
Partner
Education
Primary				−0.92**	−0.06	−1.64*
Secondary				ref	ref	ref
College				3.51***	4.08***	−0.34
Activity
Full time				ref	ref	ref
Part time				−6.55***	−4.82***	−10.17***
Inactive				−6.85***	−6.58***	−8.49***
Constant	33.56***	23.09***	54.63***	29.54***	26.49***	63.78***
R ²	.05	.04	.01	.23	.11	.10

Note: All values are weighted. ref = reference category.

p < .10. **p <.05. ***p < .01.

Results for Model 1 in Table 3 show that cohabitation yields negative coefficients for women’s contribution to all domestic chores and for cleaning. The effect for occasional tasks was not significant. By country, and taking as a reference the United Kingdom, women from all countries contribute more to overall tasks and to cleaning—especially so in Southern European countries and France, and also to occasional tasks in Italy and Germany. In France, the effect is negative for occasional tasks, and in Spain it is not significant. These effects do not disappear when the other control variables are introduced (Model 2); in fact, cohabitation shows a positive effect on women’s contribution to occasional tasks that was not significant in Model 1. The only exception is Germany, where cohabitation is no longer associated to the time spent on occasional tasks.

Working less than full time or not being in the labor market as well has having only primary education and one or more children are associated with higher involvements on total and routine tasks (although the effect of education is not significant for cleaning). Women who are college educated, those whose partner has more time available because he is not involved full time in the labor market, and those with children younger than 6 years contribute less to both total and routine tasks. Having a partner with college education reduces the contribution to total housework and cleaning. On weekends, even if there is more time available, women’s dedication to domestic chores lowers. Regarding occasional tasks, working less than full time and having primary education increase the time devoted to those chores whereas number of children decreases it.

To sum up, the effects of cohabitation in this first stage of the analysis point at cohabitors having more egalitarian distributions of housework. Cohabiting women spend significantly lower amounts of absolute time than marrieds in typical female tasks, and cohabiting men spend less time than married men in domestic work in general (models not shown). Thus, more egalitarian distributions are mainly due to women’s lower investment in housework. Cohabiting men do not increase their dedication to female tasks significantly; nor do cohabiting women spend more time on male tasks. The former models controlled for socioeconomic status, time availability, and life cycle; therefore, the remaining effects of cohabitation would be attributed to differences in values or in partner selection criteria.

On a second stage of the analysis, we wanted to test whether the effects of cohabitation differ by country using an interaction. Table 4 presents results of the model but only for the variables involved in the interaction since the coefficients for control variables did not change significantly (full models available from the author). The reference categories here are being married and living in the United Kingdom. The coefficient for cohabitation indicates in this case whether cohabitors do higher or lower proportions of the household domestic tasks than marrieds for those couples resident in the reference country, and what we observe is that cohabiting women in the United Kingdom do not differ significantly from marrieds, and only do slightly higher contributions for cleaning, although the difference is not very significant.

Table 4.

Results for the Interaction Terms of Ordinary Least Squares Regressions on Women’s Dedication to Housework, by Task.

	Total	Cleaning	Occasional
Cohabitation	0.68	2.87*	−2.68
Country
Germany	1.53**	3.08***	0.09
France	7.92***	12.76***	−8.22***
United Kingdom	ref	ref	ref
Italy	12.35***	11.76***	1.54
Spain	10.58***	10.39***	−2.68*
Interactions
Coh × Germany	−4.42**	−7.72	−1.5
Coh × France	−3.43**	−7.97***	9.96**
Coh × United Kingdom	ref	ref	ref
Coh × Italy	−1.33	−4.82***	12.59**
Coh × Spain	−4.20**	−6.74***	6.52

Note: All values are weighted. ref = reference; coh = cohabitation.

p < .10. **p < .05. ***p < .01.

The coefficient for the country of residence, however, shows the differences in housework dedication across the countries studied for the reference living arrangement, which is marriage. In this case, results for women’s percent contribution show that married women in all countries contribute more to housework—total and routine—than those in the United Kingdom, but in Spain and France they also contribute less to occasional tasks.

Finally, the interaction term in the analysis tells us whether cohabitation has different effects by country. Cohabitation decreases women’s contribution to total domestic work in Germany, Spain, and France, when compared with the United Kingdom, and it also decreases the proportion of cleaning performed in France, Italy, and Spain. Concerning occasional tasks, cohabitation increases women’s contribution to these chores both in France and Italy (the effect for Spain is also positive but not significant). Given that socioeconomic and life-stage variables were controlled in the analyses, this result supports Hypotheses 2.

Conclusions

This article has compared the gendered allocation of household labor between married and cohabiting couples in Germany, France, the United Kingdom, Italy, and Spain, analyzing harmonized time use data. Using this data source, it has been shown that for all these countries, cohabiting unions hold more egalitarian divisions of household labor, both in terms of quantity—percentage of household load—and quality—types of task performed—of that division. This egalitarianism is due to cohabiting men and women spending less time in the tasks that are traditionally attributed by gender (Hypothesis 1a), but the analysis neither finds a general trend of cohabiting men engaging more on female activities nor of cohabiting women engaging more in male activities (Hypotheses 1b). In fact, cohabiting couples simply spend less time doing domestic chores.

It was also hypothesized that, provided that cohabitation plays different roles across countries, its effect on the division of domestic work would differ too. This effect is mediated by four factors that make cohabitation selective of couples with specific traits in each country: socioeconomic status, life stage, mate selection, and values. Higher effects were expected for Southern Europe and France, because cohabitation is more selective along those factors in these cases. In the analyses, only socioeconomic status and life cycle stage were adequately controlled for; therefore, we attribute the effects of cohabitation to gender attitudes and mate selection.

This second hypothesis was explored using interactions, taking as a reference cohabitors in the United Kingdom, and the analysis provided mixed evidence. We expected the effects of cohabitation in Spain and Italy to be very similar and stronger than in other countries. In Italy, cohabitation increases men’s time investment in overall housework and specifically on routine tasks, and it decreases women’s dedication to those tasks. Consequently, cohabiting men in Italy perform a higher share of overall housework and routine tasks and also lower shares of occasional tasks. Cohabitation in Spain has similar effects than in Italy for men, but no significant effect for women, which leads to the conclusion that the egalitarian effect found for cohabitation in Spain when percent contributions were analyzed is due especially to men’s increasing contribution, whereas the effect in Italy is due both to men’s higher and women’s lower involvement in tasks. Cohabitation in France increases men’s share mainly through their own time investment, and this effect is very similar in Spain, which is not what was expected according to the meaning of cohabitation in Southern Europe. This could mean that the nature of the relationship is changing in Spain, and that it is not anymore a marginal living arrangement. No significant differences were found between cohabiting women in Germany and the United Kingdom in absolute time, but when men’s percentage contributions were examined, cohabitation in Germany is associated to higher shares of overall and routine housework than in the United Kingdom, and its effect is not lower than in other countries.

The present study has many limitations that must also be acknowledged and that could be improved by better data and by further research. One of those limitations relates to the type of data. Although time use data provide researchers with rich information and more objective measures of daily activities, the measurement of housework remains problematic (Lee & Waite, 2005). With the harmonized data, we could not explore externalization patterns, which could be a key factor in explaining distributions of domestic chores. Furthermore, harmonized data do not allow the study of very specific tasks, and the aggregate variables may be problematic, for instance we could not separate care for adults—a typically female task—from other occasional tasks, which are typically masculine.

In addition to this, the data analyzed here are cross-sectional. In order to fully understand the effects of cohabitation—and in general, to understand changes in the allocation of domestic work, it is essential to account for the biographical dimension (Bühlmann et al., 2010): it would be necessary to examine whether or how the allocation of housework changes when cohabitors marry, have children, or form a second union. Longitudinal data production is highly problematic, and especially in this case, where respondents have to fill detailed diaries. A further limitation of this study is that it lacks information on respondents’ attitudes and gender values as well as information on mate selection processes. Time use surveys do not include this information, at least not in the harmonized versions, and not only do these variables render cohabitation a selective living arrangement, but they also play an important role in the division of domestic work.

This study has shown that cohabitation has different effects across countries and has provided some insights about which factors may be relevant in order to determine the effects of cohabitation, but this represents just a first step. Cohabitation is a multifaceted living arrangement, whose meaning may evolve, and which is chosen for a variety of heterogeneous reasons: from economic uncertainty to lack of commitment or specific ideologies. Clearly, it has several meanings coexisting in the same countries, even if one of them is prevalent, and this heterogeneity should be addressed in future studies if we want to understand better the dynamics of this relationship and its role in family-formation processes.

Footnotes

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 disclosed receipt of the following financial support for the research, authorship, and/or publication of this article:

This work was supported by the Spanish Department of Science and Innovation (Grant Nos. CSO2009-11883 and CSO2010-17811).

Notes

References

Anxo

(2002). The allocation and the gender division of labor in France and Sweden. Proceedings of the France Symposium ILO, 99-108.

Batalova

Cohen

P. N.

(2002). Premarital cohabitation and housework: Couples in cross-national perspective. Journal of Marriage and Family, 64, 129-144.

Baxter

(2005). To marry or not to marry. Marital status and the household division of labor. Journal of Family Issues, 26, 300-321.

Becker

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

Bianchi

S. Z.

Milkie

M. M.

Sayer

L. C.

Robinson

J. P.

(2000). Is anyone doing the housework? Trends in the gender division of household Labor. Social Forces, 79, 191-228.

Breen

Cooke

L. P.

(2005). The persistence of the gendered division of labor. 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

Tettamani

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

Clarkberg

(1999). The price of partnering, The role of economic well-being in young adults’ first union experiences. Social Forces, 77, 945-968.

10.

Clarkberg

Stolzberg

Waite

(1995). Attitudes, values and entrance into cohabitation versus marital Unions. Social Forces, 74, 180-192.

11.

Coltrane

(2000). Research on household labor, modelling and measuring the social embededness of routine family work. Journal of Marriage and the Family, 62, 1208-1033.

12.

Davis

S. N.

Greenstein

T. N.

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

13.

Fuwa

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

14.

Geist

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

15.

Gershuny

(2000). Changing times, work and leisure in postindustrial society. New York, NY: Oxford University Press.

16.

Gershuny

Fischer

Altintas

Borkosky

Bortnik

Dosman

Yeung

(2010, January). Multinational time use study, versions World 5.5.3, 5.80 and 6.0. Oxford, England: University of Oxford. Retrieved from http://www.timeuse.org/mtus/

17.

Gershuny

Sullivan

(2003). Time use, gender, and public policy regimes, Social Politics, 10, 205-228.

18.

Gupta

(1999). The effects of transitions in marital status on men’s performance of housework. Journal of Marriage and the Family, 61, 700-711.

19.

Hallberg

Klevmarken

(2003). Time for children, a study of parents’ time allocation. Journal of Population Economics, 16, 205-226.

20.

Heuveline

Timberlake

(2004). The role of cohabitation in family formation, The United States in comparative perspective. Journal of Marriage and Family, 66, 1214-1230.

21.

Hook

J. L.

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

22.

Kiernan

K. E.

(2002). Cohabitation in Western Europe, trends, issues and implications. In Booth

Crouter

A. C.

(Eds.), Just living together, implications of cohabitation on families, children and social policy (pp. 3-31). London, England: Lawrence Erlbaum.

23.

Kiernan

Estaugh

(1993). Cohabitation, extra marital childbearing and social policy. London, England: Social Policy Studies Center.

24.

Lee

Waite

(2005). Husbands’ and wives’ time spent on housework, a comparison of measures. Journal of Marriage and Family, 67, 328-336.

25.

Lennon

M. C.

Rosenfield

(1994). Relative fairness and the division of housework. The importance of options. American Journal of Sociology, 100, 506-531.

26.

Leridon

Toulemon

(1995). Country-specific studies on the trends in family formation and the new role of women: France. In Blossfeld

H.-P.

(Ed.), The new role of women: Family formation in modern societies (pp. 77-97). Boulder, CO: Westview Press.

27.

Lück

Hofäcker

(2003). Rejection and acceptance of the male breadwinner model. Which preferences do women have under which circumstances? (Globalife Working Papers 60). Retrieved from http://www.sfi.dk/graphics/ESPAnet/papers/Hofackerluck.pdf

28.

Lundberg

Pollak

R. A.

(1996). Bargaining and distribution in marriage. Journal of Economic Perspectives, 10, 139-158.

29.

Meil Landwerlin

(2003). Las uniones de hecho en España [Cohabiting couples in Spain]. Madrid, Spain: Centro de Investigaciones Sociológicas.

30.

Meil Landwerlin

(2005). El reparto desigual del trabajo doméstico and sus efectos sobre la estabilidad de los proyectos conyugales [The unequal distribution of domestic work and its effects on marital stability projects]. Revista Española de Investigaciones Sociológicas, 111, 163-180.

31.

Ono

(2003). Women’s economic standing, marriage timing and cross-national contexts of gender. Journal of Marriage and Family, 65, 275-286.

32.

Perkins

W. H.

de Meis

(1996). Gender and family effects on the second shift domestic activity of college-educated young adults. Gender & Society, 10, 78-93.

33.

Pittman

J. F.

(1995). The effects of work history and timing of marriage on the division of household labor, a life-course perspective. Journal of Marriage and the Family, 58, 78-90.

34.

Sayer

L. C.

Bianchi

S. M.

Robinson

J. P.

(2004). Are parents investing less in children? Trends in mothers’ and fathers’ time with children. American Journal of Sociology, 110, 1-43.

35.

Schoen

Weinick

R. M.

(1993). Partner choice in marriage and cohabitation. Journal of Marriage and the Family, 55, 408-414.

36.

Shelton

John

(1996). The division of household labor. Annual Review of Sociology, 22, 299-322.

37.

Sigle-Rushton

McLanahan

(2002). The living arrangements of new unmarried mothers. Demography, 39, 415-433.

38.

South

S. J.

Spitze

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

39.

Stewart

(2009). Tobit or not tobit (IZA Discussion Paper No. 4588). Bonn, Germany: Institute for the Study of Labor. Retrieved from http://ftp.iza.org/dp4588.pdf

40.

Thompson

Walker

A. J.

(1989). Gender in families. Women and men in marriage, work and parenthood. Journal of Marriage and the Family, 51, 845-871.

41.

West

Fenstermaker

(1993). Power and the accomplishment of gender. In England

(Ed.), Theory on gender/feminism on theory (pp. 151-174). New York, NY: Aldine de Gruyter.