Introduction

During the last several decades, unprecedented changes in work and family roles have occurred in North America (Sayer et al. 2004). One of the most significant economic and social changes has been the increase in women’s participation in the workforce (Major and Germano 2006). In the United States, paid work outside the home and housework have been historically divided according to gender, where men were breadwinners who worked for pay and women were homemakers who worked at home, hence the expression “gendered allocation of labor” (Sayer et al. 2004). Research suggests that, in spite of women’s increased commitment to the labor force market and their associated political and social achievements, their advances have not been paralleled in the familial sphere (Arrighi and Maume 2000). Recent studies have shown that the gains women have made outside the home have not translated directly into an egalitarian allocation of household labor (Fuwa 2004; Fuwa and Cohen 2007; Gershuny 2000; Knudsen and Wærness 2008; Lincoln 2008). In fact, it is well documented that American women continue to perform the vast majority of unpaid tasks performed to satisfy the needs of family members or to maintain the home (e.g., Artis and Pavalko 2003; Erickson 2005; Mannino and Deutsch 2007; Pinto and Coltrane 2009). Because most of the household labor falls on their shoulders, women are typically in a relatively unfavorable position compared to men to pursue demanding career opportunities and professional advancement (see Poeschl 2008; Lothaller et al. 2009 for a similar argumentation).

The gendered allocation of household labor has become a recurrent theme of discussion, debate, and research. The current article reviews research published on the division of household labor between 2000 and 2009 (see Coltrane 2000; Shelton and John 1996, for reviews of earlier research). The aim of our review is to offer readers a comprehensive outlook on complex and ample new data and theoretical perspectives. It represents a timely effort to synthesize the available literature and to shed light on the different levels of analysis adopted by researchers as well as on the links that exist between these levels. In the applied perspective of attaining equity between men and women in daily life and given the social importance of this objective, such an examination is necessary to illuminate the processes that favor or hinder an egalitarian sharing of unpaid housework. Increased knowledge of factors favoring gender equity in the organization of daily home life could help women reach a more favorable position in both the familial and public spheres.

To complete this examination, published studies were extracted from academic databases, with “division of household labor,” “division of housework,” “heterosexual couples”, and “gender” specified as the search criteria. The reviewed research was based on probabilistic samples of individuals (with the exception of Bartley et al. 2005; Claffey and Mickelson 2009; Kroska 2004; Pinto and Coltrane 2009). We include studies based on samples restricted to married individuals (e.g., Davis and Greenstein 2004; Lincoln 2008; Mannino and Deutsch 2007; Parkman 2004) as well as those based on samples comprising cohabiting and married individuals (e.g., Davis et al. 2007; Geist 2005; Iversen and Rosenbluth 2006; Knudsen and Wærness 2008; Nooman et al. 2007).

In his article reviewing the research conducted between 1989 and 1999, Coltrane (2000) argued that scholars had achieved a better understanding of the predictors of variations in women’s participation in household labor, but were only just beginning to understand why men’s contribution to household tasks did not mirror women’s contribution. Overall, Coltrane (2000) suggested that several consistent predictors, such as women’s employment, income, and ideology as well as men’s work hours and ideology, were useful in explaining women’s share of housework. However, he noted that a number of variables used in the studies published in the 1990s had weaker and less consistent predictive value in predicting men’s involvement in household labor than in predicting women’s. Researchers who published in the following decade built upon preceding research and brought many important theoretical, methodological, and analytical advances to the field. The current article updates Coltrane’s review (2000) by documenting these advances and reviewing the new data on this recurrent theme of study. As will be seen, we now gain a better picture of what explains the gendered distribution of household labor by taking into account multiple factors related both to the individuals sharing unpaid tasks and to the context in which this sharing occurs.

We begin by summarizing the current allocation of household labor among American couples. A summary of the way in which household labor has been conceptualized and of the methodology researchers used to assess this concept is then offered. We follow with a presentation of the main theoretical perspectives adopted to explain the way partners share housework: those focusing on micro-level mechanisms (i.e., influence of individuals’ and couples’ characteristics) and those focusing on macro-level mechanisms (i.e., influence of national contexts). In a subsequent section devoted to the most recent and innovative developments in the literature on household labor, we review studies on the cross-level interactions between micro and macro factors.

In order to document the impact of micro-level factors without these latter being affected by the variations in cultural contexts, we limit our review of the micro-level perspective to only one country: the United States. As it is a forerunner in the examination of micro-level mechanisms, the choice of this reference population was a natural one. However, as researchers argued during the 2000s, the allocation of household labor can not be thoroughly understood without taking into account the cultural context in which partners evolve. In keeping with this viewpoint, the studies reported in our section on macro-level mechanisms have a cross-national scope; they compare the political, economic, and cultural contexts of up to 34 countries in which individuals negotiate the division of household labor (e.g., Batalova and Cohen 2002; Davis and Greenstein 2004; Geist 2005). Finally, the studies on the interactions between micro- and macro-level factors help qualify American data by showing that other national contexts can modify the impact of individual-level factors.

The Gendered Division of Household Labor

Today, North American women are responsible for about two thirds of routine household tasks (Claffey and Mickelson 2009; Greenstein 2000, 2009; Hook 2006; Knudsen and Wærness 2008), with cooking and cleaning being two most time-consuming tasks (see Batalova and Cohen 2002; Bartley et al. 2005; Parkman 2004, for more details on what is considered routine tasks). In addition to doing most of the housework, women are also in charge of managing, planning, and organizing these tasks (Mannino and Deutsch 2007). Fuwa and Cohen’s results (2007) showed that American women reported performing an average of 13.2 hr of household labor per week, compared to 6.6 hr per week for their spouses.

During the 2000s, a number of studies using multiple data points documented changes in the division of household labor across time (e.g., Bianchi et al. 2000; Cunningham 2007; Hook 2006). They provide valuable information on the trends and gender differentials in how couples negotiate unpaid household work in an era when women’s time is being dramatically reallocated to paid employment. A study by Bianchi and coauthors (2000), based on data collected from representative samples of adults in the United States in 1965, 1975, 1985 and 1995, examined the changes in time spent on housework by men and women. Results showed that time spent on household labor has generally declined over time. The average time spent on these tasks fell from 17.5 hr per week in 1965 to 13.7 hr per week in 1995. Over the same period, women’s and men’s hours have converged. In 1965, women performed 30 hr and men 4.9 hr of unpaid household duties per week. By contrast, women’s work hours dropped to 17.5 hr per week and men’s hours increased to 10.0 hr in 1995. While women’s housework hours decreased from one decade to the next, men’s housework hours increased between 1965, 1975, and 1985, but then leveled off until 1995. As the authors argued, it is unclear why the increase in men’s participation leveled off in the most recent period while women’s hours continued to decline. Will men continue to increase their contribution to housework over the coming decades, albeit at a slower rate, or is there a “ceiling” to the time that men will spend on housework duties (Bianchi et al. 2000)?

Conceptualization of Household Labor

Household labor has usually been conceptualized as the set of unpaid tasks performed to satisfy the needs of family members or to maintain the home and the family’s possessions. Tasks that have been included in this conceptualization of household labor are general housecleaning, meal planning, cooking, dishwashing (or loading the dishwasher), cleaning up after meals, grocery shopping, laundry (washing, ironing, and mending clothes), caring for sick family members, yard work, car maintenance and repairs, outdoor and household maintenance, taking out the garbage, paying bills, and transporting family members (e.g., Arrighi and Maume 2000; Badr and Acitelli 2008; Cunningham 2007; Lincoln 2008). Measures usually include between three and ten of these tasks, which are considered to be representative of the set of tasks to be accomplished.

It is common for scholars to distinguish different types of household tasks. Most often, household tasks are classified in the following two categories. Often referred to as the stereotypically female tasks, routine tasks are those that are on-going, nondiscretionary, and very time-consuming. They include laundry, cooking, cleaning up after meals and doing the dishes. Often referred to as the stereotypically male tasks, intermittent tasks are done only occasionally and are more flexible and less time-consuming. They include household repairs, car maintenance, and yard work (Bartley et al. 2005; Batalova and Cohen 2002; see also Badr and Acitelli 2008, for the use of a similar categorization). Most recent household labor studies have focused on routine tasks (Batalova and Cohen 2002; Pinto and Coltrane 2009; Cunningham 2007). Batalova and Cohen (2002) argued that “by focusing on the division of labor for female chores, which are routine and ongoing, we can identify the extent of egalitarianism in the sharing of household responsibilities” (p. 746). Given its lack of attention to stereotypically male tasks, one could criticize this focus for underestimating men’s actual participation in household labor. A number of authors have argued that intermittent conventionally male tasks represent quite a small portion of all work to be done around the house, as such work often requires little time to complete (e.g., taking out the garbage) or must be undertaken much less frequently (e.g., mowing the lawn or household repairs), in comparaison to routine tasks that need to be done daily (e.g., cooking and cleaning up after meals; Batalova and Cohen 2002). The results have shown that women are responsible for the bulk of the household labor even when the conceptualization incorporates intermittent tasks (Bianchi et al. 2000; Kroska 2004).

In some cases, household labor has been conceptualized as including childcare tasks (e.g., Badr and Acitelli 2008; Hook 2006), but the majority of recent studies have excluded childcare from their conceptualization (Bartley et al. 2005; Davis et al. 2007; Fuwa and Cohen 2007; Knudsen and Wærness 2008). A number of scholars have argued that household and childcare tasks should be evaluated as distinct activities, considering that their nature and predictors differ (Bianchi and Raley 2005; Coltrane and Adams 2001; Mannino and Deutsch 2007).

Methodological Issues

The examination of studies published during the last decade reveals variations in methodology, both in the measures used and in the samples employed, to assess household labor. The two issues are discussed here.

Measure

Studies published in the 2000s were conducted using almost exclusively questionnaire-based self-reported measures and/or detailed time diaries of household labor. The questionnaire-based measures are produced in paper-and-pencil or computerized format (completed either in the laboratory setting or by mail) or in interview format (conducted in person or over the phone). Respondents are asked to answer direct questions about how housework is performed in their household, either tasks in general (e.g., “On average, how many hours a week do you spend on housework, not including childcare and leisure activities?”; Davis et al. 2007; Fuwa and Cohen 2007; Knudsen and Wærness 2008) or a series of tasks (e.g., “How many hours in an average week do you/does spouse do the following: (1) cooking or meal preparation; (2) meal clean-up and dishwashing; (3) laundry, including washing, drying, and ironing clothes; (4) cleaning house; and, (5) grocery shopping”; Bianchi et al. 2000; Parkman 2004; Pinto and Coltrane 2009). Respondents are usually asked to estimate specific numbers of hours and/or relative time-use between male and female partners (Bianchi et al. 2000; Iversen and Rosenbluth 2006; Knudsen and Wærness 2008; Kroska 2004; Parkman 2004; Pinto and Coltrane 2009), but other approaches have also been used (e.g., Artis and Pavalko 2003). More specifically, relative time-use variables are obtained either from estimates of partners’ actual number of hours spent on housework (e.g., Davis and Greenstein 2004), from estimates of partners’ percentage of work completed (e.g., Claffey and Mickelson 2009) or from ordinal scales, such as (5) “The woman always does the task,” (4) “The woman usually does,” (3) “About equal,” (2) “The man usually does,” and (1) “The man always does the task” (e.g., Iversen and Rosenbluth 2006).

Even though they were used less frequently, a number of studies conducted in the last decade relied on time diary data (e.g., Bianchi et al. 2000; Gershuny and Sullivan 2003; Hook 2006). In time diary surveys, respondents are asked to report all their daily activities using a structured diary format. This approach has the advantage of forcing respondents to report primary activities usually for all 24 hr of the day. In their own words, the respondents report their activities in the order they occur. Time spent on housework is then classified into a number of categories. For instance, in Hook’s study (2006), time-use categories included time spent on routine housework (e.g., cleaning, cooking, laundry), non-routine housework (e.g., everyday purchasing and errands, home and car maintenance, care of adults and pets, gardening), and childcare (e.g., direct care and supervision, helping, playing).

Comparisons of estimates derived from questionnaire-based and time diary methods reveal that hours spent on household labor tend to be much higher using questionnaire-based measurements than using time diary measurements (Bianchi et al. 2000). This can probably be explained by the fact that respondents answering survey questions can double-count the time spent in simultaneous activities (Coltrane 2000). With time diaries, however, respondents report only their primary activities; as a result, housework that is completed simultaneously with other tasks might not be accounted for (Bianchi et al. 2000).

As for questionnaire-based measures, absolute time spent on housework by one or both partners and the proportion of household labor performed by each partner are two common variables derived from time diaries. Although relying on ratio variables is common practice, some authors have argued that it has limitations, especially because it does not provide information on the number of hours of household labor completed by each partner (Bianchi et al. 2000; Hook 2006; Knudsen and Wærness 2008).

In sum, because time diary measures make it possible to collect more precise, more valid, and more reliable data, in part due to their structured format, we believe this methodology should be favored over questionnaire-based measures in future studies. In addition, from our point of view, using both absolute involvement and ratios of involvement would lead to a more complete understanding of the gendered division of household labor.

Samples

Whether they use a questionnaire-based measure or a time diary measure, some researchers rely on data collected from both partners (Nooman et al. 2007; Parkman 2004; Pinto and Coltrane 2009), whereas others use data collected from only one member of the couple (Arrighi and Maume 2000; Fuwa and Cohen 2007; Lincoln 2008; Mannino and Deutsch 2007). A number of studies have found a self-serving bias in the reporting of household labor (Fuwa and Cohen 2007; Kamo 2000). For instance, Kamo’s findings (2000) indicated that men tend to overestimate their own contribution to household labor but that both partners have similar estimates of women’s participation. In light of this, we strongly recommend that researchers seek both partners’ participation.

Many researchers have utilized probabilistic American surveys, such as the National Survey of Families and Households (e.g., Bianchi et al. 2000; Lincoln 2008; Parkman 2004). Also, a large number of researchers, particularly those examining cross-national data, have used data from international probabilistic samples, such as the 1994 version (Batalova and Cohen 2002; Fuwa 2004; Geist 2005; Iversen and Rosenbluth 2006; Yodanis 2005) or the 2002 version (Crompton et al. 2005; Davis et al. 2007; Fuwa and Cohen 2007; Knudsen and Wærness 2008) of the International Social Survey Programme (ISSP). The ISSP 1994 collected data from 24 industrialized countries, including the United States. The ISSP 2002 contains data from 34 countries (again, including the United States) and represents the largest existing cross-national dataset on the division of household labor. In both versions, data were collected from one member for each household surveyed.

Prominent Theories

Even though it is widely recognized that gender is the strongest determinant of housework completion, justifications of this phenomenon diverge. Two main types of theoretical perspectives dominated the 2000–2009 literature on household labor allocation: the micro-level perspective, which encompasses (a) the relative resources perspective; (b) the time availability perspective; and (c) the gender ideology perspective, and the macro-level perspective. Both types of perspectives are reviewed here, starting with the more traditional micro-level perspective, which focuses on individuals’ and couples’ characteristics.

Micro-Level Perspective

Relative Resources

The relative resources perspective, also referred to as the economic exchange hypothesis or the economic dependence model, is based upon the premise that a partner’s external resources, such as income and education, grant decision-making power (Mannino and Deutsch 2007). An important underlying assumption of this perspective is that most people consider housework to be unpleasant and want to avoid it. Therefore, the more resources, and thereby power, a person has in relation to his/her spouse, the easier it should be to bargain his/her way out of routine housework (Knudsen and Wærness 2008).

Research supports the notion that wives’ contribution to family income is related to the allocation of household labor, even when holding constant other individual characteristics, such as the number of hours worked outside the home (Bianchi et al. 2000; Knudsen and Wærness 2008; Mannino and Deutsch 2007; Pinto and Coltrane 2009). Cunningham’s 31-year longitudinal study (2007) demonstrated the powerful influence of relative earnings on the change in division of household labor: Women’s relative earnings exerted a strong influence on the change over time in men’s participation in housework. The results of Parkman’s (2004) study supported the relative resources hypothesis, revealing that the spouses’ earning ratios are related to the time they spend on housework. Partners with higher earning ratios tend to complete less housework than those with lower earning ratios. In a recent study, Pinto and Coltrane (2009) found that, within the United States, women’s relative income has stronger effects on housework for women of Mexican origin than for their Anglo peers.

Relative education is related to the division of housework in a way similar to that of relative income. Bianchi and her colleagues (2000) found that couples in which the women have more education than the men have smaller gender gaps in the quantity of household labor they accomplish. According to Davis and Greenstein (2004), as women’s economic independence and high educational attainment become more prevalent, the likelihood of men performing more housework should increase. Other data have shown that men with higher levels of education contribute a higher proportion of housework than men with lower levels of education (Gershuny and Sullivan 2003). Given that socioeconomic status is generally linked to gender ideology (Cha and Thébaud 2009), a concept that will be discussed later on, the fact that couples with higher socioeconomic status have less traditional gender ideology could explain why these partners tend to share household tasks more equally.

All in all, we believe that the relative resource perspective is useful in understanding part of the allocation of housework. However, it is important to note that a number of scholars debate the linear relationship between earning differentials among partners and the allocation of housework (Bittman et al. 2003; Greenstein 2000). In fact, their studies suggest that the link between the level of economic dependence and the division of household labor is curvilinear, in the sense that the allocation of household labor is more traditional both in households where women earn less and in those where women earn more than their partners, whereas it is more equal in households where women have income approximately equal to that of their spouse. However, as mentioned earlier, other results contradict the curvilinear hypothesis and support the idea that women’s share of housework decreases when their economic dependence decreases (Fuwa 2004; Knudsen and Wærness 2008) and, in the same way, that men’s dependency is associated with a more egalitarian division of housework (Erickson 2005; Mannino and Deutsch 2007).

Despite support found for the relative resources perspective, Evertsson and Nermo (2007) have argued that this perspective remains of modest significance in illuminating the gendered division of housework because it cannot explain why women who have resources comparable to those of their partners still do most of the housework. If relative resources were more influential, then wouldn’t each partner in a family in which the man and the woman have similar resources each do about 50% of the total housework? Gupta’s study (2007) also called into question the validity of the relative resources perspective. Specifically, his analysis suggested that American women’s relative earnings do not influence the amount of housework they perform once their absolute earnings are accounted for. These results supported his previous findings (Gupta 2006) suggesting that only women’s own absolute earnings matter in the prediction of their housework hours. As explained, it might be that women with high personal incomes have more negotiating power to obtain a more equal division of labor than women with lower incomes, regardless of how much their income is in relation to their partner’s.

Time Availability

The time availability perspective posits that the amount of time each partner works outside the home influences his/her share of housework, so that partners divide household tasks according to the time they each have available (Davis et al. 2007). Individuals who spend more time on paid work have less time to spend on household labor (Artis and Pavalko 2003). Because women’s increased presence in the labor force limits the time they have available to undertake housework, the need for their partner to reconsider and reallocate the work load at home has become a major theme of discussion and research in the last decade (Robinson and Hunter 2008).

Research on the time availability perspective has focused on the impact of various dimensions related to the employment of partners (e.g., employment status, history, and hours). A study by Bianchi et al. (2000) indicated that employment status affects both partners’ involvement in housework, with full-time and part-time employed men and women performing less housework than the unemployed (see also Ciabattari 2004; Gershuny and Sullivan 2003; Nooman et al. 2007, for similar results). Cunningham’s longitudinal study indicated that husbands of women with longer employment histories perform a relatively larger portion of housework compared to the husbands of women who have accumulated less employment experience. According to this study, women’s employment status has a greater influence on the division of housework than men’s employment status.

Even though employment status is a significant predictor of the division of household labor, time spent in the workforce has a greater impact than simply employment status (Cunningham 2007). Empirical findings have shown that women’s actual hours of employment are strongly linked to the allocation of unpaid labor in the home (Artis and Pavalko 2003). In particular, the more hours a woman works, the smaller her share of the household tasks (Knudsen and Wærness 2008; Mannino and Deutsch 2007; Pinto and Coltrane 2009) and the more her partner is involved in this labor (Cunningham 2007; Kroska 2004; Nooman et al. 2007). It would seem that the increased demand for housework that occurs when women spend time outside the home encourages men’s contribution. Similarly, men who occupy jobs that do not require long hours are more likely to contribute to household tasks as compared to those with jobs that require longer hours (Arrighi and Maume 2000, Hook 2006; Pinto and Coltrane 2009).

Despite the support for the time availability hypothesis, empirical findings show that among couples in which both partners hold a job, the women assume the larger portion of household tasks (Bartley et al. 2005). In fact, even when they work as many hours as their partners at paid employment, at the end of the day, women are responsible for much of the housework (during what has been dubbed their second shift; Hochschild 1989). Lincoln’s (2008) study, based on a sample of American men and women with full-time employment (and with nearly the same amount of time spent in the workforce), showed that women put approximately 80% as much time into housework as they do into paid work, whereas men put about 40% as much time into household labor as they do into paid labor. However, Gershuny and Sullivan (2003) paint a more nuanced picture of the division of labor between American men and women by showing that when both the time spent in paid employment and unpaid domestic work are taken into account, there is a 50:50 split between partners.

It has been suggested that it is not the amount of time women spend on paid labor per se that determines the allocation of housework, but rather that the equalizing effect is due to support for gender equality gained through workplace exposure to networks that are supportive of gender equality (Bolzendahl and Myers 2004). In fact, according to results indicating that employed women tend to hold more egalitarian attitudes toward gender roles (Fan and Marini 2000), it can be argued that the impact of employment variables on the allocation of housework could be explained, in part, by individuals’ gender ideologies.

Gender Ideology

Based in socialization theories, the gender ideology perspective posits an inverse relationship between traditional gender attitudes and an egalitarian division of household labor (Davis et al. 2007). Theorists have long supported the idea that individuals are socialized into male or female gender roles, and research has consistently shown that there are persistent views regarding how women and men are expected to behave (Cunningham 2001). This perspective is based on the idea that people’s gender ideology views are situated on a continuum that ranges from traditional gender ideologies—where a strict male breadwinner/female homemaker structure is favored—to egalitarian gender ideologies—where both partners are considered equal and share the two roles more equally. It has been shown that women who hold more egalitarian attitudes are less likely than women with traditional attitudes to report performing all of the housework, whereas men who hold more egalitarian attitudes tend to behave in a more egalitarian manner and spend more time on housework than men with traditional attitudes (Arrighi and Maume 2000; Davis et al. 2007; Fuwa 2004; Knudsen and Wærness 2008; Parkman 2004).

These results converged, despite variations in how the researchers measured the concept of gender ideology. A number of researchers have measured this concept with indexes based on responses to multiple statements, such as the following: “A working mother can establish just as warm and secure a relationship with her children as a mother who does not work”; “A job is all right, but what most women really want is a home and children”; “Being a housewife is just as fulfilling as working for pay”; “A man’s job is to earn money, a woman’s job is to look after the home and family”; “All in all, family life suffers when the woman has a full-time job”; and “A preschool child is likely to suffer if his or her mother works” (see studies based on the 1994 and 2002 modules of the ISSP, such as Davis et al. 2007; Fuwa 2004; Knudsen and Wærness 2008). Other times, the concept of gender ideology was measured with single items, such as: “Is it better for the family if the husband is the principal breadwinner outside the home and the wife has primary responsibility for the home and children?” (Arrighi and Maume 2000, see also Parkman for a similar single item measurement).

Gender ideology is known to evolve (usually in the direction of egalitarianism, as each new generation tends to hold more egalitarian values than the preceding generation; Brooks and Bolzendahl 2004) and individuals tend to become more supportive of egalitarianism over their lifespan (Fan and Marini 2000). According to the gender ideology perspective, this trend toward egalitarian ideologies should translate into a more equal division of household labor between men and women. In fact, there is evidence that women who become less traditional in their gender role views over time have a diminished responsibility for housework, even after controlling for other changes that occurred during the same time in the family structure or time availability (Artis and Pavalko 2003). Similarly, the fact that younger generations of men tend to perform a greater share of the burden of household labor (Arrighi and Maume 2000; Davis and Greenstein 2004) may reflect generational differences in socialization experiences and, ipso facto, in gender ideology. Fan and Marini’s findings (2000) indicate that employed mothers hold more egalitarian gender attitudes and, correspondingly, socialize children with more egalitarian attitudes. According to these authors, such a pattern of effects, in combination with increases in education and women’s employment, suggests that the trend toward egalitarian gender ideology should continue (Fan and Marini 2000).

Overall, we believe that the gender ideology perspective is useful in explaining why the division of labor within the home remains bound to tradition despite the increases in women’s educational and career opportunities. Gender ideology is affected by education attainments and by employment (Fan and Marini 2000), therefore part of the predictive power of the aforementioned perspectives could be accounted for by the gender ideology perspective. Although studies have generally provided support for the hypothesis that holding an egalitarian ideology favors a more equal sharing of the family work, a study by Bianchi and her coauthors (2000) suggests that the gender ideology perspective is somewhat more useful in explaining women’s participation in household labor than men’s. More precisely, the findings suggest that men’s egalitarian ideology is associated with a decrease in their partners’ housework hours, but not with an increase in their own housework hours. Similarly, whereas wives with a more egalitarian gender ideology perform less housework, their egalitarian attitudes do not affect their partner’s housework hours.

A variant of the gender ideology perspective is the gender construction perspective. It suggests that the models described previously do not adequately explain the variations in the negotiation of housework because they fail to consider the gendered meanings that women and men derive from the performance of unpaid work in the home (Erickson 2005). This perspective emphasizes that the performance of housework helps define and express gender relations within the home (Bianchi et al. 2000). According to the gender construction perspective, household labor is perceived as a representation of caring and appreciation (Badr and Acitelli 2008), and caring for the family in the home is more central to the female role (Knudsen and Wærness 2008). Therefore, women may perform more housework because it enables them to behave consistently with their feminine and expressive gender identities and men may resist doing more housework to protect and reinforce their identities as men (Erickson 2005). In accordance with this line of reasoning, Bianchi et al. (2000) argued that women become more invested in the completion and supervision of household labor because “the cleanliness of one’s home is a reflection on women’s competence as a ‘wife and mother’—but not men’s competence as a ‘husband and father’” (p. 195).

Many scholars have turned to the gender construction perspective to explain results regarding the gendered allocation of labor. A study by Arrighi and Maume (2000) showed that when American men face challenges to their masculinity in the workplace, they are more avoidant of housework. The authors argued that this resistance to contribute to housework is important for men in maintaining their masculine identities. Davis and Greenstein (2004) have formulated two possible explanations for the observation that men’s ability to use their relative resources to negotiate out of housework is greater than that of women. Both explanations are in concordance with the gender-construction perspective: 1) It is easier for men to use their relative power to obtain a more favorable division of household labor because of the low expectation for men’s contribution to traditionally female tasks; and 2) women might choose not to use their relative power to negotiate out of the housework in order to protect their feminine identity.

Overall, the literature reveals that the distribution of household labor is influenced by multiple forces. Relative resources, time availability, and gender ideology all prove to be important predictors of the gap between men’s and women’s unpaid labor. However, we believe no perspective yet offers a clear explanation of why women still do the bulk of housework even when they display the personal characteristics that favor a more egalitarian division of household labor. The literature review leads us to conclude that the division of household tasks is a complex process that may be best understood by relying on a combination of theoretical perspectives.

Macro-Level Perspective

In the 2000s, more and more scholars argued that macro-level factors are as important as micro-level factors in the understanding of the division of housework between partners (Fuwa 2004; Geist 2005; Hook 2006). Recently, researchers have argued that, to arrive at a more complete understanding of the human social world, it is crucial to give more consideration to the social context in which the behaviors occur (e.g., Davis and Greenstein 2004). The macro-level perspective is based on the idea that structural and cultural forces shape the way individuals behave in their own home and, particularly, how they organize and share housework. In that regard, throughout the last decade, research on the allocation of unpaid labor has begun to focus on how national contexts (e.g., political, economic, cultural contexts) influence the negotiation over housework and on how these contexts can be modified through political efforts and social policies (Batalova and Cohen 2002; Fuwa 2004; Fuwa and Cohen 2007; Geist 2005).

A number of authors have documented the importance of national contexts in the understanding of partners’ behaviors in the home. Along the same lines of reasoning as those who were interested in the effects of individuals’ gender ideology, researchers have recently studied the role played by gender equality at the macro-level. The Gender Empowerment Measure (GEM; United Nations Development Program 2004) has been used to evaluate macro-level gender equality, or female empowerment, in many countries (Batalova and Cohen 2002; Fuwa 2004; Knudsen and Wærness 2008). This measure is an indicator of women’s professional opportunities, economic power, and participation in politics. The national GEM scores are based on the percentages of parliamentary seats occupied by women, the percentage of female administrators and managers, the proportion of female professional and technical workers and women’s share of earned income, in comparison with that of men. Overall, results from studies based on 22 (Batalova and Cohen 2002; Fuwa 2004) and 34 countries (Knudsen and Wærness 2008) suggested that couples in more gender-egalitarian societies (e.g., Canada, Sweden, the United States) tend to divide housework more equally than those living in less gender-egalitarian societies (e.g., Chile, Italy, Japan), even when holding constant their individual characteristics. Women in the most gender-equal nations perform an average of 15 hr of housework per week, whereas those in the less equal nations perform around 27 hr of housework per week (Knudsen and Wærnesss 2008).

Evidence has been found that the impact of women’s position in the society goes beyond individual negotiations within the couple and affects the general population. For instance, Hook’s results (2006) indicated that not only married but also single men spend more time doing household tasks in countries where women’s employment is more common. According to this author, the fact that men are more involved in the private sphere when women are more present in the public sphere could be explained by societal shifts in gendered roles and not merely by bargaining that occurs between partners.

Researchers have also been interested in the way in which political context influences interactions between men and women in intimate relationships. Geist (2005) found that it is particularly rare for couples to share household labor equally in countries with conservative welfare-state regimes that are described as actively supporting traditional gender roles (e.g., Japan, Italy, Austria). In contrast, the results of her study indicated that social-democratic regimes that actively advocate gender equity (e.g., Sweden, Norway) have higher levels of equal sharing. Results concerning liberal regimes that are focused on individuality (e.g., Australia, Canada, the United States) are more heterogeneous. However, the dominant pattern is one where women are primarily responsible for housework, though equal sharing is more common in the United States and Canada than in the other liberal regimes. These patterns of results are independent of individual characteristics (i.e., gender ideology, time availability, or relative income).

Pursuing this notion further on the basis that political context is malleable, people have proposed that political efforts could shape the national context and, in turn, influence not only the publicly visible forms of gender equality (e.g., in the workforce and in the political sphere) but also the less visible behaviors that occur in the privacy of the home. Through the study of state policies, scholars have sought to extend their understanding of the connections that exist between social policies, national context, and individuals’ behaviors. In general, researchers have hypothesized that social policies in areas such as work regulation, work–family balance, and gender equality initiatives could influence the gendered allocation of household labor. For instance, it has been proposed that work regulations limiting employed individuals’ standard working time could increase men’s availability to complete unpaid work in the home (Gornick and Meyers 2003; Hook 2006; Jacobs and Gerson 2004). It has been argued that publicly funded childcare policies that support the work–family balance could help women negotiate a fair division of household labor by facilitating employment and financial independence, thereby reducing their entrenchment in the caregiver/homemaker role (Hook 2006). In contrast, it has been anticipated that parental leave, because it is used primarily by mothers, could hinder a more egalitarian division of labor by reinforcing the gendered male breadwinner/female homemaker roles in addition to limiting women’s financial resources and long-term employment possibilities (Deven and Moss 2002; Hook 2006). Finally, it has been proposed that, by promoting women’s participation in the workforce and their financial independence, gender equality initiatives could give women more bargaining power to negotiate a more equal division of labor (Fuwa and Cohen 2007). Despite being somewhat inconsistent, the available data concerning the direct effects of social policies on the allocation of household labor between men and women offer some support to the previous assumptions.

For instance, with regard to the impact of state policies aimed at facilitating the work–family balance, Fuwa and Cohen’s study (2007) documenting the influence of parental leave policies on the division of household labor in 33 countries, including the United States, yielded interesting results. Their analyses indicated that the length of available parental leave is associated with the division of labor between partners. Couples in countries that offer generous parental leave policies (e.g., Slovakia) tend towards a more egalitarian division of household labor than couples in countries with shorter parental leaves (e.g., Chile), but exceptions were noted (e.g., the United States). Hook’s results (2006), based on a cross-national sample from 20 countries counting the United States and spanning over 35 years, nuanced this finding by specifying that it is supported only for couples in countries where men are eligible to take parental leaves. More specifically, in countries that extend access to parental leaves to men, fathers completed about two extra hours of housework per week compared to their counterparts in countries where men are not eligible for that benefit. However, when all countries (including those offering only maternity leave) are taken into account, Hook’s results suggest that lengthy parental leaves tend to facilitate a traditional division of labor between spouses. The conclusion was that extending parental leave eligibility to both parents reduces the support for specialization and the adherence to traditional gender roles.

With regard to the impact of affirmative-action policies and public childcare on the division of household labor, recent studies have yielded inconsistent results. Through a comparative analysis of dual-earner couples in different welfare states, namely Britain and France, Windebank (2001) verified the impact of policies concerning women’s employment and childcare provision on the gendered division of household labor. Results suggest that state policies favoring women’s employment and facilitating childcare arrangements do not bring about improvements in the allocation of household labor between men and women. These findings led the authors to conclude that “it remains the mother’s, not the state’s, and not the father’s, responsibility to organise and conduct domestic and parenting work” (Windebank 2001, p. 286). The data actually suggested that publicly funded childcare could limit fathers’ contribution by maintaining attitudes about fathers’ uninvolved role. It appears that, by taking over some of the functions that typically fall into women’s hands (such as childcare), the state does not necessarily encourage men to become more involved (see Kitterød and Pettersen 2006, for a similar argument). Windebank (2001) proposed that a more equal sharing of household and parenting tasks might be more likely in contexts where men are forced to partake in childcare out of financial necessity or lack of alternatives. On the other hand, other results paint a somewhat more positive picture concerning policies that favor women’s employment and suggest that these latter can influence women’s bargaining power on the private level. More specifically, a cross-national study by Iversen and Rosenbluth (2006), based on data from 14 countries, including Canada, Japan, Sweden and the United States, indicated that the division of household labor is more egalitarian in countries with friendly labor markets and large public sectors (e.g., Canada) than in countries where the opposite is observed (e.g., Japan). In addition, national levels of employment opportunities for women explained an important proportion of the cross-country variation in the gendered allocation of household labor.

In short, at this time in research, there is some evidence supporting the idea that political efforts and social policies can modify the behaviors that occur in the home, but results are inconsistent. As Crompton and her coauthors (2005) argue, inconsistent results may be explained by the fact that the impact of social policies is countered by other simultaneously occurring social changes. Through a cross-national study conducted in two waves, these scholars aimed to document the changes in the gendered allocation of household labor in three rather different countries (i.e., Britain, Norway, and the Czech Republic). It was observed that, despite the contrasting state policies that differentiated the nations under study, men’s involvement in household work remained roughly the same in all three countries. The authors concluded that, although national social policies supporting the work–family balance are increasingly common, increases in pressure and competitiveness in the workplace might have worked against government efforts by making men’s participation in housework more difficult. Based on qualitative and quantitative data, Lewis and Smithson (2006) arrived at the conclusion that policies to support gender equity can help, but are undermined by lack of change in the workplace. The persistence of a traditional ideal of full-time and uninterrupted work, intensified in the context of global economic trends, can mean that fathers who take advantage of entitlement programs (for instance, parental leaves) are undervalued in the workplace and may see their careers limited, just like many mothers who work part time (Lewis and Smithson 2006).

Overall, the body of results obtained to date might be disappointing for policymakers wishing to influence gendered behaviors and attitudes. In this regard, we believe that the impact of social policies may be lessened by the fact that they are frequently reversed with each change of government and, when maintained, that they take time to mature and be accepted by the population (see Crompton et al. 2005, for a similar argument). We agree with Bernhardt et al. (2009) who recently argued that policies need to be in place for some time before they can influence individuals’ ideals and actual behaviors regarding gender equality in the home.

Cross-Level Interactions

Various authors have recently argued that national-level variations in gender equality (for instance, in wages, career opportunities, or political presence) could play a moderator role in the relationship between micro-level factors (relative resources, time availability, and gender ideology) and the division of household labor. Fuwa (2004) reasoned that women in countries where social conditions are more egalitarian may have less difficulty using their individual-level characteristics (e.g., income, work hours, gender role ideology) to negotiate a fair division of household labor, whereas women’s income, work schedule, or values might not have as much influence on the allocation of household tasks in countries where the gender inequalities are relatively more present in the social sphere (see Knudsen and Wærness 2008, for a similar argument). Research has generally supported the idea that high national-level gender inequality tends to dampen the impact of individual-level factors (e.g., Fuwa 2004; Knudsen and Wærness 2008).

The equalizing effects of women’s full-time employment and of individuals’ egalitarian gender attitudes on the allocation of household tasks within couples were found to be stronger in countries where women hold more power in general (Fuwa 2004). Similarly, Geist’s study (2005) documented that, for women, less traditional gender ideology is associated with a more egalitarian division of housework in liberal (e.g., Canada, the United States) and social-democratic (e.g., Norway) regimes, but this effect is not present in conservative regimes (e.g., Japan). In contrast, this same study indicated that women in countries with conservative regimes experience a greater reduction in housework hours for each additional hour of work spent in paid employment, in comparison with their counterparts in countries with other regimes. The author concluded that in conservative countries, women may need to display concrete behaviors, such as working more hours, which decrease their availability around the house, in order to gain a more equal sharing of household tasks. This is because, unlike in social-democratic and liberal countries, symbolic aspects, such as their attitudes and values, are not sufficient to modify the allocation of labor in the home. Overall, it appears that without a reduction in macro-level gender inequalities, gains made at the individual level may not be sufficient to achieve a more egalitarian division of household labor.

In the same vein, the results of Fuwa and Cohen (2007) suggest that social policies have a moderating effect on the associations between women’s assets and the division of housework between partners. More specifically, the authors found evidence that the equalizing effect of women’s full-time employment on housework is stronger in countries with affirmative action policies favoring women’s employment (e.g., Australia, the United States) and weaker in those with longer parental leave policies (e.g., France, Slovakia). In the same way, the effect of women’s higher relative income was found to be stronger in countries without discriminatory policy limiting women’s employment opportunities (e.g., Australia, the United States).

Knudsen and Wærness (2008) have investigated the interplay between national context and individual characteristics in influencing not only partners’ relative contributions to household labor (i.e., the proportion of housework completed by each partner within a couple) but also their absolute contributions (i.e., the number of hours of housework completed by each partner individually). This inclusion of absolute contributions has added a nuance to work done previously (e.g., Fuwa 2004; Fuwa and Cohen 2007; Geist 2005), which was mainly focused on relative contributions between spouses. On the one hand, their results did support the idea that micro- and macro-level factors interact to modify the partners’ relative contribution. More specifically, the findings suggested that, in gender-egalitarian countries (e.g., Norway, Sweden, the United States), higher income and less traditional gender values increase the woman’s power to achieve a more egalitarian division of household labor, whereas these individual characteristics have little influence on the relative contribution of two partners living under national conditions of less female empowerment (e.g., Chile, Russia). On the other hand, this same micro-level/macro-level interaction was found to be non-significant in analyses focusing on each spouse’s absolute contribution. When examining absolute contributions instead of relative contributions, the results revealed that both the micro- and macro-levels have an impact on the hours of housework completed by each partner individually, but the impact of one level does not vary according to the other level. More precisely, women with higher relative income, less time availability, and less traditional values spend fewer hours completing household tasks no matter what country they live in and, similarly, women in egalitarian countries spend less time in household labor no matter their relative income, time availability, and gender ideology.

In short, the results obtained to date point to an interplay of macro-level and micro-level forces on partners’ contribution to household labor. Overall, we are convinced by the data showing that national levels of gender equality do influence women’s ability to use gains made at the individual level to negotiate a more equal division of housework. We believe that these results constitute important advances in our understanding of the gendered division of household labor. However, from our point of view, much research remains to be done given that little is known about the causal mechanisms by which the macro-level factors infiltrate the micro-level negotiation of unpaid work within the home.

Conclusion

In summary, this review made it clear that in the United States household labor remains persistently divided along traditionally gendered lines. Women continue to assume the larger portion of household chores. Although changes toward more gender equality have been noted across cohorts, very few scholars hold an optimistic vision. In fact, a number of authors express reserves about the convergence of men’s and women’s time spent on household tasks and perceive that changes in men’s contribution to household labor occur at a slow pace (e.g., Bianchi et al. 2000; Cunningham 2007). Considering that unequal family practices limit women’s availability and opportunities to take on a greater role in the professional, social, and political spheres (Poeschl 2008), it is crucial that we refine our understanding of the underlying meanings and purposes given to household labor that help perpetuate men’s and women’s gendered allocation.

In the past decade, American researchers have pursued work on the causal factors that help explain the variations in household labor allocation observed among couples. The current paper sheds light on the new data supporting the traditionally dominating perspectives ,which focus on micro-level characteristics: women with greater relative incomes and educational attainments, as well as those who are less available to complete work at home due to paid employment and those who favor more egalitarian gender attitudes, tend to share household labor more equally with their partners (Arrighi and Maume 2000; Artis and Pavalko 2003; Bianchi et al. 2000; Cunningham 2007; Mannino and Deutsch 2007; Parkman 2004). Despite their usefulness in explaining part of the phenomenon, these perspectives do not offer a clear explanation of why women often continue to be responsible for completing most of the housework even when they display the individual characteristics that favor a more equal sharing (Bittman et al. 2003; Greenstein 2000).

Throughout the last decade, various findings have supported the idea that researchers need to move away from a focus that is put solely on the individuals’ and couples’ characteristics to one that considers the broader social context in which they evolve (Davis and Greenstein 2004; Hook 2006; Knudsen and Wærness 2008). Through their macro-level perspective, researchers demonstrated that national contexts in which female empowerment is greater and in which political structures advocate gender equity tend to favor egalitarian allocations of household labor within couples (Batalova and Cohen 2002; Fuwa 2004; Knudsen and Wærness 2008). Researchers have also found some evidence that political efforts enacted by the establishment of social policies can sometimes modify the allocation of household labor that takes place in the home (Fuwa and Cohen 2007; Hook 2006; Iversen and Rosenbluth 2006). More research remains necessary to understand the long-term impacts of these social policies and disentangle the effects of political efforts and of other social changes that co-occur. This will help policymakers achieve a better understanding of which set of social policies could bring about more equity on the individual level within their particular national context. Here, it is important to underscore that studying the impact of social policies is challenging, considering that they are in constant evolution, that they vary greatly from one country to another, and are subject to considerable revisions and even abolishment with each changing government.

Studying the impact of micro- and macro-level factors independently has its limitations, particularly because the two types of mechanisms tend to be interconnected. In fact, research on micro-level factors can be interpreted in light of macro-level factors (e.g., individuals’ gender ideology is related to the dominant value system of the society in which they evolve, just as time availability is connected to employment and social policies). This has led researchers to study the interaction between both types of variables (Fuwa 2004; Fuwa and Cohen 2007; Knudsen and Wærness 2008). Building upon previous work, researchers’ recent investigations have revealed that macro-level factors can actually moderate the impact of a number of individual characteristics on the way household labor is shared within couples (Fuwa and Cohen 2007). Specifically, as revealed by the current paper, we now know that gains made by women at the individual level are more likely to translate into a more equal division of household labor with their partners in countries where social conditions are more egalitarian (Knudsen and Wærness 2008).

In conclusion, research clearly indicates that how couples share the housework in their home is shaped through complex processes. The recent advances in research stress the importance of understanding the multifaceted ways in which individual characteristics as well as social, cultural, and political contexts shape individuals’ behaviors and influence couples’ relationships. It appears quite clear that we need to continue studying individuals within their broader social context if we wish to arrive at a thorough understanding of the persistent gendered division of household labor and what could change it.

We believe that to refine our research approach, the conceptualization of household labor should be re-evaluated. If we wish to gain more knowledge of the differences and similarities in unpaid work that exist between men and women, we believe a conceptualization that includes a broader range of tasks should be adopted. If all productive activities completed outside of paid employment are to be taken into account, intermittent and childcare tasks, emotion work, volunteering, and support work should be studied alongside the routine tasks usually measured, instead of in separate literature (see Erickson 2005; Hook 2004; Minnotte et al. 2007, for similar arguments). Furthermore, in the future, we should follow in the footsteps of the growing number of researchers who study both absolute and relative contributions to unpaid labor. Researchers should also consider broadening their focus to the way all men and women, not only those in intimate heterosexual relationships, manage these tasks (see Kurdek 2007 for a sample of the work that is developing in this area).

Between 2000 and 2009, few qualitative studies have attempted to explain the asymmetry in the division of household labor. A number of studies used a qualitative methodology, but were focused mainly on participants’ perceptions of fairness and unfairness regarding the allocation of household labor within their home (see Mannino & Deutsch 2007, for an example) and on the impact of the division of household labor on couples’ relationships (e.g., Klein et al. 2007). To gain a better understanding of the meanings that people attach to their division of labor patterns, more qualitative research needs to be conducted.

In comparison with researchers working during the 1989–1999 decade (see Coltrane 2000, for a review), researchers working in the 2000s were no longer looking for the theory (among those that have been developed) to explain the complex phenomenon of the distribution of household tasks across genders. Indeed, they now share an increasingly global view of the situation and generally concur that a multitude of factors come together to maintain the traditional distribution and that micro and macro theories all have a role to play. The simultaneous study of individual and social factors has nuanced prior studies that dealt solely with one or another type of variable. Despite this progress, work is still needed to fully comprehend which conditions are necessary if men and women are to share unpaid housework more equally and how to favor these circumstances.