Gendered Division of Domestic Labour and Childbearing Intentions in Tehran, Iran.

Objective: Changes in the gendered division of domestic labour are often assumed to influence childbearing intention, but existing evidence is varied and less examined in the Asian context. This paper aims to investigate the association between the gendered division of domestic labour and the intention to have another child. Materials and methods: A cross-sectional survey was conducted in Tehran on 455 married women aged 18-40 years who were selected through a multi-stage cluster sampling.
Results: Domestic labour is still a feminine role and the majority of women are satisfied with the division of domestic labour. Women's satisfaction with the division of domestic labour is a predictor of their tendency to have another child, but the actual division of domestic labour has not a significant effect on women's desire to childbearing, in the multivariate model.
Conclusion: Women's desire for having another child is positively impacted by their satisfaction with the gendered division of labour in their household. To achieve more fertility, gender equality in the family and identifying the factors affecting women's satisfaction with the division of domestic labour is suggested.

Introduction

Gender equality is the concept that women and men have equal conditions, treatment and opportunities for realizing their full potential, human rights and dignity, and for contributing to (and benefitting from economic, social, cultural and political development. Gender equity is the process of being fair to men and women, and importantly the equality of outcome and results. Equity ensures that women and men have an equal chance, not only at the starting point but also when reaching the finishing line (1). Domestic labour as one of the indicators of gender equality in the home has been traditionally handled by women and decision-making on family affairs with men (2). Although many changes have occurred in the family function, structure, and couples relationships in recent decades, scholars point out that gendered division of domestic labour has remained relatively stable in many countries (3-6).

During recent decades, the total fertility rate has dropped below replacement level in many developed countries (3, 5, 6). Although a large body of researches has focused on economic motives, opportunity costs of having children, economic uncertainty and shifts in ideology, gender equality has been less posited in understanding the low fertility rate (7-9).

The association between domestic labour and childbearing, following the rise of female labour force participation and women’s double burden was considered. Different views exist as to whether divisions of domestic labour and men’s greater involvement in domestic labour will make people wish for more or fewer children. Some scholars mentioned that increasing gender equality had been linked to a low fertility rate for a long time (10, 11). However, recent studies have reached conflicting results, By increasing gender equality in the family which in its turn gives women relief from housework, they will achieve a higher chance to combine their job responsibilities with their domestic duties, which is positively associated with recovering childbirth (12, 13). In a different finding, Torr and Short indicated that there is a U-shape relation between the women’s share of housework and the number of children in the United States (13).

Since the actual impact of gendered divisions of domestic labour and egalitarian values on fertility is unclear, understanding demographic prospects in the 21st century is crucial (14), especially in developing societies with low levels of gender equality and fertility rates. There is a lack of sufficient empirical evidence in Asian countries where women may face particular challenges combining their career and family responsibilities due to the cultural context (5). Some studies have shown that there is still great variation in family arrangements, which means many women may be perceived as fair even when the actual division is unequal (15). Therefore, in the relationship between fertility and the division of domestic labour, satisfaction may be a more important factor than actual domestic labour (10, 16, 17).

In the case of Iran, although women have had equal access to higher educations, their attendance in labour market has meaningfully remained low. The dominant economic pattern in Iranian families has been the male-bread winner model; so employed women have faced difficulties combining their responsibilities in their workplaces and their maternal duties at home. As a result, men’s participation in domestic labour could impact women’s fertility intentions. On the other hand, Iran is a country with moderate levels of development, changes in various dimensions of the family life (18) presence of some traditional attitudes especially in men (19), low level of women’s labour force participation, and a low fertility rates (20).

Family power structure is a key notion mentioned by feminist scientists. They claim that distribution of power within the household and mainly power imbalances between husbands and wives plays an important role in their reproductive decision- making (21).

Goldscheider et al. declare that gender revolution took place in two-stage. At first, women participation in activities outside their homes such as their attendance in labour market rose; it was at the beginning of that phase when fertility intention decreased among women who were in employment compared to those were not employed. It could be explained by the fact that employment did not release women from their obligations at home. Indeed, after hours of working, employed women had to start their work in the household. In spite of that double burden on women, fertility rate went up and got closer to the replacement level in the second phase due to increase of men participation in domestic labour (22).

McDonald suggests that the decline in fertility rate is linked to combination of high levels of gender equity in individual‐oriented institutions, as education and employment, and low levels of gender equity in the family and family‐oriented institutions. He states that countries like Netherland, Denmark, Finland and Sweden with high level of gender equity, experiences the high level of fertility around 1.8 as well. On the other hand, countries with low level of gender equality such as Italy, Spain and Greece faces with lower level of fertility rare which is around 1.3 (23).

Theory of gender equality claims that, women’s new opportunities in education and labor market raise perceptions of unfairness when they are not supported. It is really hard for women to be mothers and go for higher education or take part in labour market at the same time. Employed women are not treated fairly and experience more injustices if they become mothers. This negatively impacts on their fertility decision-making. In the other word, low fertility rate occurs in the region where women’s progress in their workplace be in contrast with household work while they suffers from lack of equality and enough supports at homes (7).

Nowadays women have more contribution in the family income; nonetheless there is still a mass of household chores remain on their shoulders. Some scholars believe that women’s perception of fairness regarding to their share of domestic labours is more substantial than the actual division of domestic labour (23). Division of domestic labour may be unequal, but it is perceived as fair, it may be related to increased childbearing (3, 17, 24). So, the way gender equality is defined by spouses and their understanding of fair division of domestic duties should not be considered as less important as actual division of labour itself. Therefore, women’s childbearing behaviour is not necessarily in contrast to the actual amount of house work done by them; but it is somehow related to women’s opinion about the fairness of their share of domestic labour.

Various studies look at the situation differently. Some researchers indicate that an unequal division of household chores is associated with a decreased chance of first and subsequent births (3, 14, 22, 25-27). However, some scholars claimed there is no significant relationship between gender equality and childbearing (28). Finally, some studies found the relationship between domestic labour and fertility patterns to be insignificant (29).

Iran is an important case for analysis in this area for several reasons. Iran, as one of the Asian countries, has experienced decreasing fertility rate in a short period. Since 2006, the fertility rate in Iran has dropped below the replacement level (2.1 children per woman). The decline in childbearing in Iran is more similar to developed countries than to Asian countries with similar development levels. Meanwhile, Tehran has had low fertility rate for more than two decades and is currently a city with the lowest- low fertility rate (1.2 children) (30).

Iran has experienced rapid several socio-cultural changes in the family structure and women’ role in the household (18). Some researches on Iranian family report different changes such as increasing divorce rates, decreasing desire for childbearing, extramarital sex, change in emotional life of families, cohabitation, differences and sometimes generational conflict (31). An emphasis on individual autonomy, privacy, and self-fulfillment is increasingly developing in couple’s relationship (18, 32); while others suggested that extreme individualism is not dominant in spousal relationship (19).

Although there are a low level of women’s labour force participation (17/6%) (20), university education is widely seen as another source of power. The share of women in postgraduate education has risen from 31% in 1997 to more than 60% in 2018. In addition, human development index in Iran was 0.798 in 2018 and ranked 60 among 189 countries (33). Finally, in spite of government policies in order to increase population, the birth rate remained low; and the impact of gender equality in the family on childbearing intentions has been less considered. In such a context, women’s childbearing behavior and their preference to have offspring could be related to their spouses’ behaviors in terms of the gender division of domestic labour. Therefore, considering the combination of tradition and modernity in various aspects of Iranian family lifestyle, the relation between gender equality and fertility is ambiguous. The aim of this study is to investigate, whether actual division of domestic labour and the satisfaction of domestic labour division, have any impact on the intention to have another child. This article take women with different childbearing transitions, including those with no child, one child and two children. In this regards, two following hypotheses is introduced:

Hypothesis 1: There is a positive association between actual division of domestic labour and intention of having another child.

Hypothesis 2: There is a positive association between division of domestic labour satisfaction and intention of having another child.

Materials and methods

We used data from a cross-sectional survey on "Married Lifestyle and its determinants" that was conducted by the first author of this paper in Tehran, the capital of Iran, in 2015 (19). The statistical population was married men and women in Tehran and the sample size was considered to be 1728 individuals. So that, in accordance with Cochran’s formula we could keep our sampling error at 0.05% and our confidence level was 95%. In addition, we considered the design effect at 2 and an adjustment factor of 0.25% in case of missing responses. We increased the number of people in the sample up to 2000, due to the possible loss of participation which is calculated to be around 0.15 based on the family census of the Statistical Centre of Iran. In each household, the husband or wife was invited for the interview. We used the probability proportional to size sampling method. The project manager evaluated all the distributed questionnaires and those questionnaires which largely remained unanswered were put aside. Finally, we did our ultimate analysis based on the number of 1736 questionnaires. Experts and faculty members of the institute modified Questionnaires to gain face validity. Additionally, exploratory factor analysis was applied to obtain construct validity and Cronbach’s alpha was used to check the reliability of the items.

In this analysis, 455 married women aged 18-40 years from 50 districts of Tehran were selected. Women over 40 were excluded because naturally both fertility intention and actual fertility rates are very low among them. The operational definitions of the variables are as follows:

The dependent variable was measured by asking one question: Do you intend to have another child? The response was coded as 1 when the respondent intended to have another child and 0 when the respondent did not intend to have another child.

Equality in domestic labour is based on the division of five different tasks: Who is responsible in your family for? (1) Cooking and preparing daily meals, (2) Dishes washing, (3) Food shopping, (4) Doing small repairs in and around the house, and (5) House cleaning. The responses for these items were categorized into: always the woman, usually the woman, woman and man equally do it, usually the man, and always the man. Based on the results of factor analysis, two items of food shopping and doing small repairs in and around the house were excluded. Internal consistency for the scale was good (α=0.75). Based on the index score, respondents were classified into two groups of “women more” and “men equal and more”.

This is represented by satisfaction, which is in line with prior research such as Buber (3) and Neyer et al. (17). This variable is measured using one item: How satisfied are you with the division of household tasks between you and your spouse? The scale for this variable ranged from 0 to 5, where 5 denoted being very satisfied with the division. The score of 4-5 was categorized as high, 3 as moderate and 0-2 as low satisfaction.

Age is measured into 3 groups; 18- 29, 30-35, and 36-40 years old.

Total number and sex composition of surviving children.

The education level of respondents is defined in three categories: Lower secondary, higher secondary, and tertiary education.

The employment status of respondents is defined in two groups: not being employed, and being employed.

Family monthly income is measured in three categories; less than 600 US $; 600-1000 US $; and more than 1000 US $.

According to the Huber study (34), religiosity was measured by a five-item scale in which the participants rated each of the statements using a 6-point Likert-type that the designed scale was ranged from 0 (not at all) to 5 (too much). These items included: (1) To what extent do you believe in an afterlife, (2) How often do you pray? (3) How often do you take part in a religious community? (4) How often do you experience situations in which you have the feeling that God or something divine is present? And (5) How interested are you in learning more about religious topics? Internal consistency for the scale was 0.90. Based on the index score, respondents were classified into three groups of religiosity; “low”, “moderate” and “high”.

According to Swim et al. (35) and Glick et al. (36), gender role attitudes were measured by a 7-items scale in which the participants rated each of the following statements using a 5-point Likert-type designed scale ranging from (strongly disagree) to 4 (strongly agree). These items are: (1) Managerial positions for women is acceptable, (2) Men are better political leaders than women, (3) University education is more important for men than for women, (4) Women should not be employed if her husband supports her, (5) Woman’s duty is housekeeping and child care, (6) The man must be the main decision-maker in the family, and (7) Husband should earn a larger salary compared with his wife. Internal consistency for the scale was good (α=0.89). Responses were reversely coded so that the higher the score is, the more traditional gender role attitudes. Based on the index score respondents were classified into three groups of “egalitarian”, “intermediate”, and “traditionalist” attitudes. Since in this sample less than 5% were in the traditional group, we reduced the three groups to two groups; less traditional and more traditional.

Our model included two main explanatory variables: division of domestic labour, and Satisfaction with the division of domestic labour. Multivariate analysis using binary logistic regression is conducted to assess the association between the division of domestic labour and intention to have another child controlling for other variables. Also childless, one-child and two-child women were analyzed separately.

Results

The distribution of socio-demographic characteristics of samples is displayed in Table 1. The results indicated that around 37% of the sample was aged 30-35 years old. More than 50% of the respondents had tertiary education. Women's employment was 28%. About 45% of the samples had income less than 600 US $ in a month. Almost 48% of the respondents were categorized as very religious, and half of them had less traditional gender role attitudes. About 20% of women in our sample had both sexes of children, 26% only son and 25% only daughter.

Table 1

Intention to have another child by the number of living children

Number of living children	Intention to have another child		Total
	No	Yes	Percent (number)
0	18.0	82.0	100 (133)
1	46.3	53.7	100 (164)
2	90.2	9.8	100 (123)
3+	100.0	0.0	100 (35)
All	54.1	45.9	100 (455)
Statistical test	Cramer's V= 0.608, Sig.= 0.001

Considering the division of domestic labour, 91% of women were more responsible and only 9% of men were equally/or more responsible for domestic labour. In addition, about 80% of women were satisfied with the division of domestic labour in a moderate and high levels. Almost 93% of women were satisfied with the division of domestic labour when men equal/ or more than women were involved, and in the case that women were more involved in domestic labour, 79% of them were satisfied with division of domestic labour.

As it is shown in Table 2, 29.2% of the respondents were childless, 36% had one child, 27% had two children and 7.8% had three and more children at the time of the survey. In addition, 46% of all the respondents, 82% of childless women, 54% of women with one child, and 10% of women who had two children liked to have another child.

Table 2

Socio- demographic characteristics of the respondents by the number of living children

Variables		No children	1 child	2 children	All
Division of domestic labour	Men equal and more	17.3	9.8	0.8	9.0
	Women more	82.7	90.2	99.2	91.0
Satisfaction with division of domestic labour	Low	16.5	20.1	22	19.8
	Moderate	26.3	44.5	42.3	38
	High	57.1	35.4	35.8	42.2
Age (3)	18-29	46.6	32.9	8.9	27.9
	30-35	42.9	34.1	35.8	36.3
	36-40	10.5	32.9	55.3	35.8
Education	Lower secondary	3.0	10.4	23.6	15.2
	Higher secondary	15.8	36.0	48.8	33.6
	Tertiary	81.2	53.6	27.6	51.2
Family income (monthly)	<600 $	26.5	45.7	57.4	45.4
	600-1000 $	24.2	22.2	24.6	23.2
	> 1000 $	49.2	32.1	18.0	31.4
Employment	Non-employed	61.7	69.5	82.1	72.1
	Employed	38.3	30.5	17.9	27.9
Religiosity	Low	46.6	20.1	3.3	22.0
	Moderate	34.6	31.1	26.8	30.3
	High	18.8	48.8	69.9	47.7
Gender attitude	Less Traditional	72.9	52.4	30.1	50.1
	More Traditional	27.1	47.6	69.9	49.9
Composition of living children	Without children	100.0			29.2
	Only son		50.6	24.4	25.7
	Only daughter		49.4	20.3	24.8
	Both sexes			55.3	20.2
Sample size (n)		133	164	123	455

The results of multivariate analysis are displayed in Table 3 that shows the intention to have another child by the predictors. These predictors can explain up to 64% of the variance of the dependent variable. Only 3% of women without children had low secondary education, so lower and higher secondary education levels have been merged for childless women. Moreover, only 0.8% of men in the households with two children, participate in domestic labour as equal as or more than women.

Table 3

Logistic Regression Coefficients (Standard Error): Intention to have another child by the number of children at the time of interview

Variables		No child	1 child	2 children	All
Division of domestic labour (ref. Women More)		1.264 (0.730)	0.082 (0.777)		0.537 (0.494)
Satisfaction with domestic labour (ref. High)	Low	-2.488 (0.784)*	-0.204 (0.620)	-2.799 (1.293)*	-1.502 (0.426)*
	Moderate	-1.235 (0.774)	0.373 (0.531)	-3.992 (1.204)*	-0.898 (0.362)*
Age (ref. 36-40 year)	18-29	2.495 (0.895)*	2.977 (0.583)*	2.458 (1.292)*	2.702 (0.418)*
	30-35	1.071 (0.787)	1.823 (0.502)*	1.916 (0.928)*	1.582 (0.356)*
Education (ref. Tertiary)	Lower secondary	0.611 (0.848)	0.147 (0.774)	-1.986 (1.332)	0.061 (0.545)
	Higher secondary		-1.051 (0.507)*	-1.032 (0.931)	-0.719 (0.358)*
Family income (ref. > 1000 $)	<600 $	0.583 (0.991)	0.420 (0.796)	2.274 (1.605)	0.495 (0.513)
	600-1000 $	1.647 (0.969)	-0.195 (0.790)	1.689 (1.645)	0.388 (0.499)
Employment (ref. Non-employed)		-0.354 (0.735)	-1.014 (0.709)		-0.422 (0.457)
Religiosity (ref. High)	Low	-1.853 (1.072)	-1.111 (0.635)	-0.145 (0.974)	-1.021 (0.439)*
	Moderate	-0.387 (1.133)	-0.587 (0.491)		-0.430 (0.366)
Gender role attitudes (ref. More traditional)		-0.218 (0.874)	-1.428 (0.486)*	-2.725 (1.143)*	-1.397 (0.370)*
Composition of living children (ref. both sexes)	Without children				-1.906 (1.233)
	Only son		-0.465 (0.416)	-1.124 (1.118)	-0.664 (0.700)
	Only daughter			-0.136 (1.134)	-0/222 (0.690)
Number of living children					-3.070 (0.587)*
Constant		0.023 (1.928)	0.643 (0.823)	-1.773 (1.485)	3.772 (1.353)*
Model ummary	Cox & Snell R Square	0.239	0.351	0.213	0.478
	Nagelkerke R Square	0.390	0.468	0.449	0.639
Hosmer and Lemeshow test	Chi-square	12.840	7.957	2.358	1.503
	Sig.	0.117	0.438	0.968	0.993

P<0.05

Also, there is no employed woman who wants another child. Therefore, these two variables were excluded from this model. In addition, only 3.3% of women with two children were categorized in low religiosity, therefore they have been merged into moderate religiosity.

Based on the results in Table 3, there is no significant association between the husband’s involvement in domestic labour and fertility desire (hypothesis1), although according to the bivariate test may be significant in case the sample size increase. There is a significant correlation between the satisfaction of women with the division of domestic labour and fertility desire (hypothesis2). Thus, by controlling other variables, the intention to have another child will increase significantly if women satisfied with the division of domestic labour.

Moreover, the intention to have another child among younger women who are not employed and less educated and lower income was higher. In addition, those who were more religious and those who had traditional gender attitudes were more intended to have another child in all samples and transition to one child, two children and three children.

We also tested the interaction between variables in the models. In the last model, the interaction between male participation in domestic labour and religiosity was controlled and there was almost no difference in results. In the model of childless women, the interaction between satisfaction with the division of domestic labour and employment had a strong meaningful effect. So that dissatisfaction in employed women had the most negative impact on them choosing not to be a mother.

Discussion

In this study, 45.9% of the samples had a desire to have another child. Although 34.8% had already had two or more children, 29.2% were without children and 36% had only one child. The actual number of children for the women aged between 18 and 29 in the sample was 0.60 on average. 81.1% of these women intended another child. 1.02 children on average are recorded for other respondents included in the age group of 30 to 35. Our data indicated that nearly half of the women in this group, 50.3%, desired to have another child. The rest of the women were classified in the third group and their ages ranged between 36 and 40. They had the highest number of children on average which was 1.71 per woman. On the contrary, there were just 14.1% of these women had intentions of having another child. As literature says “Couples in Iran tend to have two children, specifically one son and one daughter” (37).

Indeed, no significant association between the husband’s involvement in domestic labour and fertility desire is detected. And the significant correlation between satisfaction of women with the division of domestic labour and fertility desire is confirmed. Thus satisfaction of women with the division of domestic labour was a stronger predictor to have another child compared to the actual division of domestic labour. Overall, in our study, fertility desire had little to do with the gendered practices at home between husbands and wives. But the significance of women’s satisfaction with the division of domestic labour cannot be ignored.

Child number and age were critical factors. The intention to have another child decreased sharply with increasing age and number of living children. Religiosity increase childbearing intention, in contrast, egalitarian gender role attitudes decrease intention to have another child.

In this study, we found that domestic labour is still a feminine task; and gender inequality in the division of domestic labour is substantial. In line with the first stage of the gender revolution, women gained new opportunities to go for higher education, be employed and have great participation in family decision-making, but a traditional pattern expecting women to be more responsible to domestic labour is still persistent. However, an unequal division of domestic labour leads to low fertility.

These findings confirm the theory of gender equality. This theory claims that women’s new opportunities such as education and occupation, raise perceptions of unfairness when women are not supported in their families. This is particularly happens when women become mothers as they did not receive enough support; so that, they might prefer to have fewer or no children.

These findings are in line with studies in western and Asian countries, such as Myrskylä et al. (27), Kim (5), Dommermuth et al. (9), Alonso (38), Tazi-Preve and at al. (39), Cooke (40), Kan and Hertog (25).

Despite clear evidence that gender inequalities exist in the division of domestic labour between the two sexes, most men and women regard this division as fair. In this study, we found this fact that the majority of women do the most share of the domestic labour in their families, even when they are full-time employed. It seems an unequal gender division in Tehran has been accepted as a social norm. Because, despite the large inequality in domestic labour, most women are satisfied with the division of domestic labour at a moderate level or higher. This means there are still great variations in family arrangements in Tehran, which may be perceived as fair and satisfactory even when the actual division is not equal. It is also essential to consider that there is already a large burden placed on the ‘male-breadwinner 'family structure, particularly in these family-based societies.

We are faced with increases in women’s tendency for higher education, participation in the labor market, more income and so on; on the other hand, reduction of traditional gender attitudes and changes in religiosity. Therefore, other affective variables, such as the relationship between spouses, should be emphasized and the proposed policies should be considered these factors.

Existing theories, including the theory of feminists, claim that the first stage of the gender revolution, mainly emphasizes the decline in women's childbearing in unequal conditions. According to the first stage of the gender revolution, employed women have less inclination to have children due to their double burden, but this research shows that in the dual-worker families men also want fewer children (18). Therefore, in future studies, we should focus on men and their fertility desires.

As we told, the interaction between satisfaction with the division of domestic labour and employment had a strong meaningful effect. So that dissatisfaction in employed women had the most negative impact on them choosing not to be a mother. Given the rising trend of women's employment, it is important to pay attention to this issue.

The intention of having children is far more complicated than what we tried to indicate in our explanatory model. We had a finite set of data and were deprived of longitudinal data. The analyses also were relied on a survey in which only the wife or husband in the household was interviewed, which means that it is uncertain whether the partners have the same perceptions of the domestic labour division. Finally, fertility intentions were discussed only to women’s and not couples’ intentions. Women do not make fertility decisions in a vacuum. Future studies would be to examine differences in the childbearing intentions of the couple. Also, future researches should investigate the causal effects of the division of domestic labour on fertility by employing longitudinal data.

Conclusion

Women’s desire for having another child is positively impacted by their satisfaction with the gendered division of labour in their household. To achieve more fertility, gender equality in the family and identifying the factors affecting women's satisfaction with the division of domestic labour is suggested.