[Gender division of domestic work in the Spanish municipalities]
Altuzarra Artola, Amaia
Gálvez Gálvez, Catalina
González Flores, Ana María
Department of Applied Economics V
University of the Basque Country (UPV/EHU)
Área Temática: (Mercado de trabajo y territorio)
Resumen:
The main aim of this work is to empirically examine the differences in the relationship between own and their partners´ relative and absolute incomes and the amount of time spend on childcare and household chores of Spanish couples across municipalities of different size. We distinguish five types of municipalities according to the size: more than 100.000 inhabitants, between 50.000-100.000 inhabitants, between 20.000 and 50.000 inhabitants, between 10.000-20.000 inhabitants and less than 10.000 inhabitants.
The data come from the latest Spanish Time Use Survey (2009-2010). By doing this, this work makes a contribution to gender literature regarding the household division of housework and childcare by examining the role of the Gupta’s (2007) hypothesis based on the relevance of women´s absolute earnings versus the traditional models that support the idea of the importance of relative earnings on the distribution of domestic work.
Our findings show differences among sexes and across municipality sizes. Regarding the time spent on household chores, results for large municipalities show that time devoted to household chores decreases with women income. In addition, we find that when women´s absolute earnings equal or surpass those of their partners, they do more housework in order to respond to their socially expected gender role. As for childcare, results shows that in large municipalities, the time spent by women on caring for their children remains constant until women contributes to half of the household incomes, from that point on, the time spend on childcare decreases.
The estimates presented seem to fit better the largest municipalities (over 100.000 inhabitants) and for the smallest ones (less than 10.000 inhabitants). It would be desirable to have a larger number of household in each group of municipalities in order to have results more robust. Notwithstanding, our findings suggests that the associations between (absolute and relative) incomes and time spend on housework differ across municipalities´ sizes.
Palabras Clave:gender, time-use, relative incomes, absolute incomes, Spain
Clasificación JEL:C8, J22, J16
1. INTRODUCTIÓN
The rapid incorporation of women in the labour market over the last recent decades has increased their total workload. As counterpart, they have gained economic autonomy and intra-household negotiation power. In this new context, the carrying out of domestic duties in dual-earner households has become a challenge for the partners since tasks that traditionally were performed by women must be distributed between both partners. Evidence shows, that in comparison with the recent past, total time allocated to domestic work has decreased considerably mainly due to the lower time allocated to unpaid work by women (Bianchi et al. 2000). However and even though men contribution to domestic duties has increased over the time, women still support the highest share of these tasks (Sayer 2005). In other words, the increase in women’s labour force participation has not been accompanied by a proportional decrease in the time they devote to domestic work and men’s contribution to domestic tasks is still modest. Therefore, the household specialization pattern has not changed significantly (see, for example, Gershuny 2000 and Bianchi, et al., 2006; Sevilla-Sanz et al., 2010).
This persistence in the pattern of distribution of domestic work within the household has important policy implications for gender equity (Lachance-Grzela and Bouchard 2010). On the one hand, as women spend higher time on domestic work (compared to men) they have less human capital accumulation (Becker 1985) and less promotion opportunities. On the other hand, this pattern of time distribution may lead to a negative effect on women´s wages (for a review of the literature on this issue, see Hersch and Stratton 2000 and Sevilla-Sanz etal., 2010).
Furthermore, considering the total number of hours devoted to unpaid work of men and women may hide other gender inequalities. That is, not all domestic tasks are perceive as equal, some activities are seem more unpleasant than others: household chores are usually less desirable than caring for children. Therefore, it is relevant to distinguish between these two tasks since the distribution of them between the partners may influence the women´s quality of life.
Three main strands of literature have explained the division of domestic work between women and men. All this lines of thought share the idea that the distribution of unpaid tasks is related to the level of resources (usually, earnings) that each partner contributes to the household. Briefly, the first strand defends a negative association between women´s relative earnings (compared to their partners) and the time they spend in domestic work. Human capital (Becker 1991) and exchange bargaining (Brines 1994) theories are within this line of reasoning. Human capital view suggests that different investment in human capital of women and men explain men´s higher earnings and, their greater time spent in paid work compared to women who invest less in human capital and consequently spend more time in unpaid domestic work. Bargaining approach suggests that women and men negotiate over domestic work, so that the partner with higher earnings (usually men) is in better position to bargain and can decide to do less unpaid work. The second strand is known as the gender display, or gender deviance neutralization approach (Brines 1994; Greenstein 2000; Bittman et al. 2003) which suggest that as women’s relative earnings equal or surpass those of their partners, men tend to compensate for their gender atypical household earnings contribution by doing less housework, while women compensate by doing more housework. The third strand proposed by Gupta (2006, 2007) argues that it is women’s absolute earnings, rather than their relative earnings, that determine how much time they spend on housework. He shows that women’s relative earnings do not influence their housework hours when simultaneously controlling for their own absolute earnings (Baxter and Hewitt, 2013)
Empirical results so far are inconclusive and depend on the country and/or the period analysed. For example, some studies for United States and Australia have found support for a negative association between women´s relative earnings and their dedication to domestic work (Brines 1994; Presser 1994; Bittman, et al. 2003). On the contrary, Gupta (2007) found, for United States, that women´s absolute earnings determine their time allocation to domestic work.
Our hypothesis is that the socio-economic, labour, cultural and institutional context may favour a specific association between (absolute and relative) earnings and the division of domestic work between partners. There are several studies that focus in the relationship between incomes and the distribution of time spent on domestic work for different countries. However, to the best of our knowledge, there are not studies that address this issue taking into account the size of the municipalities within one country. Differences in the services provided by local authorities, commuting times, transport services, proximity of relatives, types of occupations, etc might influence the distribution of time within the members of the couple.
The main aim of this work is to empirically examine the differences in the relationship between own and their partners´ relative and absolute incomes and the amount of time spend on childcare and household chores of Spanish couples across municipalities of different size. The data used come from the latest Spanish Time Use Survey (2009-2010). By doing this, this work contributes to gender literature regarding the household division of housework and childcare by examining the role of the Gupta’s (2007) hypothesis based on the relevance of women´s absolute earnings versus the traditional models that support the idea of the importance of relative earnings on the distribution of domestic work.
This study also contributes to the literature in the following ways. First, it adds to current empirical knowledge on the role of earnings (relative and absolute) by looking at men and women separately and distinguishing between associations with men’s and women’s own characteristics (education level, domestic service, work conditions or day of the week) occupational status) and those of their partner. Second, it examines associations between the number and age of children and the domestic time allocation of each partner in dual earner households. Third, it also looks at the connections between the day of the week and the division of time within household members. Forth, it looks at the associations between women and men (relative and absolute) earnings and the time allocated unpaid work differentiating between household chores and childcare.
The remaining of the paper is organized as follows. Section 2 presents a literature review about the relationships between incomes and time devoted to childcare and housework. Section 3 describes the data source used and present some basic statistics of the variables used in the empirical work. In the fourth section the econometric strategy is explain. Section 5 shows and interprets the results. Finally, Section 6 discusses and concludes the paper.
2. REVIEW OF THE LITERATURE
Research on the division of domestic work within couples have shown that, in general, women spend more time doing household chores and child care than men, despite the fact that their income may be higher or equal to that of their husbands. Around this question, different theoretical developments have been proposed.
The theory of human capital, based on neoclassical theory, argues that comparative advantages between men and women influence the division of work (paid and unpaid) between them. This approach considers men and women as actors who collaborate to achieve a common goal: the welfare of the household and its members. It suggests that women and men invest differently in human capital, specializing in household tasks or market production according to their levels of productivity, skill, and expected income in each area (Becker, 1991). The greater the comparative advantage, reflected through higher educational level or income, the lesser the time they spend on household chores. Since traditionally men receive higher wages than women do (who invest less in human capital), the latter ones are who ultimately take care of the household and childcare. This approach concludes that the time spent by women on housework (or childcare), compared to men, decreases as their relative salary increases and, therefore, it is heavily influenced by their dedication to paid work.
An alternative theory is the relative resources perspective, also known as the bargaining approach, economic exchange or economic dependence. This theoretical framework postulates that the distribution of domestic work within the household is a reflection of power relationships between men and women. This theory, also of a neoclassical nature, points out that the relative level of socioeconomic resources provided by each household member determines the power in the negotiation process on the time spent on unpaid work (McElroy and Horney, 1981). If unpaid activities are considered undesirable by both spouses because they are usually seen as unpleasant or low-level tasks, the partner with higher relative incomes will have greater bargaining power when it comes to distributing household tasks (Brines, 1994; Greenstain, 2000; Evertsson and Nermo, 2004). The consequence of this model is that women, who in general have less access to the labour market (and lower wages) or are economically dependent on their partner, have less bargaining power and therefore are responsible for most domestic work. According to this perspective, the financial contribution of the spouses to the household income can influence their bargaining power, which is perceived as a fair exchange within the couple. If both spouses spend the same amount of hours doing paid work but one of them earns more, it may be perceived as "fair" or "appropriate" for the breadwinner to do fewer homework assignments.
In both perspectives, it is assumed that financial resources of the couple affect the time spent on domestic tasks by each member. Both theories suggest that the spouse with higher incomes will devote relatively less time to domestic duties, underscoring the relevance of the gender wage gap in determining the time spent in the household chores. However, neither human capital theory nor relative resource theory can explain why a woman who works full-time and earning as much or more than her husband continues to do most of the household work. The gender perspective explains this fact.
The gender view (also referred as doing gender or deviance neutralization hypothesis) emphasizes the weight of tradition and cultural and institutional factors. Social norms attribute roles to women and men in the division of domestic work, which are related to an identity of femininity and masculinity (West & Zimmermand, 1987; Gupta, 1999; Bittman et al., 2003; Sevilla-Sanz et al., 2010). This approach considers that household tasks are a symbolic representation of gender relations and that the distribution of domestic work is not exclusively the result of simple negotiation processes. On the contrary, the existence of traditional roles between genders makes women feel that domestic work is their responsibility and assume a greater share of household responsibilities, even when their earnings are similar or greater than their husband’s is. However, men do not have that sense of responsibility.
From this point of view, when men earn less than women, couples respond to this "gender deviance" by reaffirming their socially expected gender roles. This gender compensation is the opposite of what the exchange theory predicts. Individuals use domestic work to assert their gender identity against atypical gender economic circumstances, so that the partner with unusually high or low incomes for their gender compensates this fact with increasing their participation in the household responsibilities. More precisely, women whose income exceeds that of their husbands will spend more time in domestic work than other women, and men whose incomes are unusually low compared to those of their wives spend less time on domestic work than other men. In these cases, women spend more time doing household chores (or men less) to show that they are still women (or men) (West and Zimmerman, 1987). Insofar as beliefs and ideology influence the allocation of time to household tasks, economic efficiency decreases.
Some researchers have suggested a combination of the negotiation and gender approaches to explain the distribution of domestic tasks within couples (Brines 1994; Greenstein 2000; Bittman et al., 2003). Brines (1994) called this model "compensatory gender display". According to this theory, the amount of hours spent by wives on domestic work will decrease as they bring in a larger share of incomes (relative resources theory), to the point where they contribute to half of the household incomes. However, as wives' earnings share increases beyond this point, their time devoted to domestic will increase (gender approach).
An explanation for this phenomenon is that couples with "relative earnings deviated by gender", that is, where women earn more than their partners, will compensate this deviation by adopting a traditional division of domestic work (doing gender). The main implication of this approach is that women earning more than their husbands do not use the advantage of having greater financial resources to gain greater equality in the distribution of domestic tasks. Instead, they are penalized for their success in the paid work, by doing more household chores than they would have to do if their incomes were relatively lower.
In contrast to these theories about the distribution of domestic work, the autonomy perspective developed by Gupta is found. This approach criticizes both negotiation and gender postulates. Gupta (2006, 2007) argues that the time that women spend in the household duties is determined by their own absolute income, neither by their husband's incomes, nor by their relative income. This approach predicts a decline in the time women spend on domestic work as their own incomes rise (Gupta and Stratton, 2010, Killewald and Gough, 2010).
3. DATA AND DESCRIPTIVE ANALYSIS
3.1. DATA
We use data from the last wave of the Spanish Time Use Survey (TUS) for the period 2009-2010 elaborate by the Spanish Statistical Institute. The target population of the survey is composed of households living in main family dwellings and the household members aged 10 years old and above. The total number of households surveyed is 11.538.
This survey uses the time use diary method, which is considered the most suitable approach to gather the use of time (Craig y Bittman, 2008). All household members aged 10 years old and above must complete the activity diary on a chosen day. The diary's time sheet covers 24 consecutive hours (from 6am to 6am the next day) and is divided into 10-minute intervals. In each of these, the informant should note down the main activity, the secondary activity carried out at the same time (if applicable) and if they are with other persons at the time.