From Parental Home to the First Partner: Does Having Lived Solo Have an Impact upon the Current Division of Housework?

Ariane Pailhé, INED
Anne Solaz, INED

Living independently before partnership has become a more common step in life course trajectories. This study analyzes whether this period of independence influences later life, and especially the spouses’ division of household labor. During this period of independence, youth people learn to live on their own; young women and men have to perform household tasks. Living independently may also act upon attitudes regarding appropriate wife and husbands behaviors. We expect that living independently before partnership reduces gender differences in housework once in a relationship. We use the first wave of the French Gender and Generation survey with a sample made of 1,297 men and 1,748 women aged 20-50 in their first partnership. The spouses’ division of household labor is measured by a synthetic indicator of female over-implication based on eight common domestic tasks. We estimate OLS and 2SLS regressions to take into account a possible selection effect according to which people with more traditional attitudes are less likely to live independently and more likely to enter directly into marriage (or cohabitation) after leaving parental home. Results show that the gender division of domestic tasks is lower when men have lived independently. It is not a selection effect.

From parental home to thefirst partner: Does having lived solo have an impact upon the current divisionof housework?

Extended abstract

Littleresearch has focused on the gender division of labourin a life course perspective. Some studies have shown that it varies over thelife cycle (Anxo et al. 2011; Lam et al. 2012). Otherstudies have focused on some key transitions of the life course, such as transitionfrom cohabitation to marriage, birth of a child or couple breakdown. As far aswe know, there are no study which links early life course events and divisionof housework. Since leaving parental home and young singlehood are key markersof the life course (Schwanitz, 2017), we will focuson that stage. Time devoted to housework is likely to increase after leavingparental home. But what is the impact of living independenltybefore partnership on the gender division of labouronce in partnership?

Twoalternative hypotheses can be formulated. First, living independently afterliving parental home may be an apprenticeship period during which young peoplelearn how to do domestic activities. So it may increase their futurecomparative advantage in doing these tasks, especially for men who are lesslikely to participate in housework (Wight et al., 2009) and less likely toperform “female oriented tasks” during adolescence (Solaz& Wolff 2015). Living independently may also act upon attitudes regardingappropriate wife and husbands behaviors. In other words, people may be lesstraditional after the experience of living alone. If behaviors are in agreementwith attitudes, one can expect different effect by gender: a lower implicationof women and a higher implication of men.

Onecan argue there could be a selection into young adult singlehood: men and womenwith more traditional attitudes are more likely to enter directly into marriageor cohabitation. Thus, it is not the experience of living independently but thisselection that would play on the division of labor.

Data and definitions

Weuse the French GGS data and restrict our sample to men and women aged 20 to 50and living in their first partnership. We define the duration of young adultsinglehood as the period between parental home departure and first union. 42%of men and 56% of women have not lived independently before their firstpartnership. About 20% have spent between 3 months and 2 years independently,and about 40% of men and 25 of women have lived on their own more than 2 years.

To define the gender division ofhousework, we use the questions regarding who in the household most often carryout the task, asked for eight tasks. So, we have an indicator ranging fromminus 16 to 16.

Models

Ina first step we run a simple OLS regression on the indicator of femaleover-implication. Our interest variable is the duration of independent living beforepartnership, in discrete or continuous alternative specifications. We controlfor age at living parental home, education level, spouses employment status,union cohort, number of children, whether there is a paid help for domesticchores, household income, the share of men’s earnings in household income andfinally we control for the presence of the partner during  the interview.

Results

Resultsshow a negative effect of men’s young adult singlehood on femaleover-implication, whatever the specification, but no effect of women’s one. Asthis negative effect may come from a selection effect, we run a 2SLS model. Resultsare similar: the negative effect of young adult singlehood on femaleover-implication is still there, and there nosignificant effect for women. Other variables act as expected: femaleover-implication decreases with men’s level of education, increases when womendo not work, with union duration and when the man’s household income increases.

Table 1: OLS and IV regression:Effect of Young adult singlehood

MEN

WOMEN

 + income

 IV

 + income

 IV

< 3 months (ref.)

3 months - 2 years

-0.630**

-0.327

(0.304)

(0.244)

> 2 years

-0.607**

-0.259

(0.286)

(0.245)

Number of years

 -0.057*

 -0.063**

-0.395**

-0.0321

 -0.031

-0.208

(0.032)

(0.032)

(0.159)

(0.032)

(0.033)

(0.186)

Conclusion

Thegender division of domestic tasks is lower when men have lived independently, aresult not driven by a selection effect. In itself, living independently likelychanges men’s attitudes-it changes their view regarding men’s roles within the family-, and changeseffectively their behaviours -they do more. By increasing men’s comparativeadvantages in housework, the gender division is less unequal. There is noeffect of having lived independently for women. The independence they gainedduring young adult singlehood does not change their bargaining power within thefamily. Their bargaining power much more depends on their current employmentsituation and income.

The next step is to analyse whether the sameeffect are observed in countries with different context of leaving parentalhome. 

Presented in Session 1139: Life Course