Analyzing the Impact of Living in Joint Families on Female Labor Force Participation
by Aakriti Jain
Ashoka University
1. Abstract
Despite rapid economic growth, female labor force participation (FLFP) has been dismal in India. Burden of household chores and childcare are often cited as reasons for the same. Using OLS and IV estimation, this study establishes a negative correlation between joint family and FLFP. The results are similar across different types of employment. Further, the study also suggests a greater role of patriarch in influencing the characteristic correlation between joint family and FLFP.
2. Introduction
It has been widely observed that during the 90s when the Indian economy bloomed, it’s labor force participation (LFP) remained low. The low LFP and WFP (work force participation) have been persistent characteristic of the Indian labor markets and have also to some extent, been barriers to realization of the demographic dividend (JJ Thomas (2012)). Low female labor force participation (FLFP) has often been cited as the reason for overall low LFP. The transfer of women’s work from the household to commercial employment is a crucial and deterministic factor for economic development. However, India’s performance in this aspect has been strikingly dismal pushing its overall LFP to bottom ranks in comparison to rest of the world. According to National Sample Survey Office’s (NSSO) survey of unemployment in 2011–12, only 23% of women, which is less than one-fourth of adult women population, were engaged in paid work or reported as being available for work opportunities. Past studies have suggested that both demand side and supply side issues contribute to this low FLFP. Demand side issues like lack of opportunities, female friendly work environment and supply side issues like rising family income, longer duration required for attainment of education etc. are cited as key reasons for decline in FLFP. In fact, many studies (Sathar and Desai (2000); Das and Desai (2003)) have established a negative, U-shaped relationship between education and FLFP, i.e. as as education rises, FLFP has been puzzlingly declining.
In this blog, I aim to analyze the effect of household structure, in particular, living in a joint family on FLFP. Domestic duties like cooking, cleaning, childcare etc. are opportunity cost to female’s time. Joint families could play a role in sharing this load and enabling a greater number of females to be involved in economic activities. But, on the other hand, joint family structure may impose restrictions on the decision making ability and mobility of females, making it harder for them to be involved in formal employment.
One may argue that living in a joint family setup enables females to have access to pooled income and time which should facilitate their transition from households to employment in economic activities. Greater income and resource pool enables them to acquire better skills and increase their chances of employability. Mother-in-law, sister-in-law, aunts and other members of the family also extend a helping hand towards household chores and child rearing. Infact, for many career women, it is the joint family setup that aids them in managing full-time jobs. But, this also has chances of increasing male dominance and thereby reducing autonomy of females in decision making (Dhanraj (2019)). It is true that living together enables people to share household responsibilities, but, often working females are viewed from the lens of status symbol (Eswaran, Ramaswami and Wadhwa (2013)). Allowing the females to work reduces the perceived social status and thus, despite having the time and resources, females may restrain to participate in formal employment. In the light of this these clear trade-offs arising from residing in a joint family setup, this blog examines the role of household structure on FLFP.
3. Literature Review
International studies like, Sasaki, M. (2002) find a significant positive impact of living in a joint family on FLFP in Japan. This suggests that co-residence allows married females to share the burden of household work and increases their likelihood of performing economic activities outside the household. Greenwood and Santos (2016) also established role of changing family structure, from joint families to nuclear families as one of the many possible reasons for decline in FLFP in China. However, in context of India many researches have observed a negative correlation between joint families and FLFP. Debnath (2015) used an instrumental variable approach to estimates the effect of joint versus nuclear household structure in India on the autonomy of women and their labour force participation. After controlling for the heterogeneity in the effects by income, caste, and region, they find that women living in nuclear households have greater decision-making power. Dyson and Moore (1983) also find that the heterogeneity in female autonomy across northern and southern parts of India can be largely explained by local-level kinship. Klasen and Pieters (2015) does analysis of five large cross-sectional micro surveys and identify various reasons for stagnation of FLFP. One major reason as identified by them is the rising family income, husband’s education and family structure. Dhanraj (2019) also investigates the role of joint family structure on non farm FLFP and finds a similar negative impact. Borrowing from the existing literature, this blog studies the impact of joint family on employment status of female in the formal and informal sector as well as on self employment.
4. Data & Variables
This blog exploits the Indian Household and District Survey (IHDS). Jointly organized by researchers from the University of Maryland and the National Council of Applied Economic Research (NCAER), it is a nationally representative, multi-topic panel survey of 41,554 households in 1503 villages and 971 urban neighborhoods across India. The first round of interviews took place in 2004–5. A second round of IHDS, re-interviewed most of these households in 2011–12 with a 83% re-interview rate. It covers information about topics concerning health, education, employment, economic status, marriage, fertility, gender relations, infrastructure, wage levels etc. In this study, I present, a cross-sectional analysis of impact of living in a joint family on FLFP using a sample of married females aged 15–60 by merging individual, household and eligible woman level data from IHDS-II.
This blog creates an indicator variable for female labor force participation. In the present study any female who worked for at least 240 hours in the previous year across all types of work except caring for household animals and collection of firewood is considered to be in the labor force. Caring for household animals and collection of firewood has been excluded because these normal household chores which do not qualify as economic activities. The Labor force participation has further been desegregated into three categories, namely, self employment, salaried employment and casual employment. If the female lives in a household where there is more than one married male or female, she is said to be living in a joint family. Education is split into 5 categories, uneducated, primary, secondary, higher-secondary and graduate or higher. Similarly, Age has also been split into 6 categories : 15–19, 20–24, 25–34, 35–44, 45–53, 54–60. Similar manipulations are done for household head’s age and education. Caste, religion, region (urban or rural), other household income inclusive of NREGA and pension benefits, number of children, number of sick days(being frequently ill, limits the ability of the female to participate in economic activities), disability index, and state and district fixed effects have also been taken into account. The table below presents the summary statistics for the variables used.
5. Econometric Analysis
5.1 Ordinary Least Squares
To test the hypothesis of what kind of effect, if any, joint family has on FLFP, the following OLS regression equation has been used:
Here, Yi is an indicator variable of whether a particular female ‘i’ is employed or not. Xi represents if the particular female lives in a joint family. As mentioned earlier, a female is said to be living in a joint family if she resides in a household with more than one married male or female. So, β1 is the main coefficient of interest depicting the effect of joint family on FLFP. Ci is a vector of individual and household control variables including the number of days the respondent has been ill, disability(if any), age, education, availability of household help, religion, caste, region, sex of household head, head’s education and age, no. of children born, alive and presently living with the respondent, other family income including NREGA and pension benefits. epsilon is the error term. State and district fixed effects have also been controlled for. Further, the blog uses four variations of the dependent variable (Yi) to desegregate the effect of joint family on FLFP by type of employment. The outcome variable of interest in the first variant is whether a female is employed at all or not, in any kind of work. In the second variant, if the female was employed in own farm or family business, she is considered to be self employed. In the third variant, if she is involved in the formal sector, she is considered to be a salaried employee. And finally in the fourth variant, if she is a daily wage worker in either farm or non farm activities, she is considered to be employed casually. The main results of interest are presented in table 1.
The regression results suggest a small but negative impact of joint family on FLFP. If the likelihood of residing in a joint family goes up by 1 unit, then, ceteris paribus, on average, the likelihood of the female being employed in any kind of work reduces by 0.05 units. Apart from self employment, these results are significant at conventional levels at the aggregate as well as dis-aggregated level, which stands in complete contrast to the popular belief that joint family eases female’s household responsibility and thereby enables her to work outside the household. These estimates however, maybe biased due to the endogeneity problem. Females who wish to work may choose to not get married into a joint family. To solve this problem, I use an instrumental variable approach.
5.2 Instrumental Variable
Two instrumental variables (IVs) for joint family: Father-in-law (FIL) still alive and Mother-in-law (MIL) still alive have been used in this analysis. For the IV to be a valid one, two conditions must be satisfied. One, the instrumental variable should be uncorrelated with the error term, this is called the exclusion restriction condition. Two, after controlling for other independent variables, the IV should be partially and strongly correlated with the endogenous explanatory variable which is joint family in our case. If the correlation between the IV and the endogenous explanatory variable is not strong, the IV is called a weak IV. While the second condition can be tested for, due to the unobservability of the error term, the first condition is difficult to verify. So, one has to rely on economic theory and natural experiments for the satisfaction of this condition. The paper borrows the reasoning for using FIL still alive from Debnath (2015) who stresses the role of a male household head in influencing the co-residence of different generations. Further, the paper borrow the reasoning for using MIL still alive from Khanna and Pandey (2020) who emphasize that a co-residing MIL may restrict women’s labor force participation as the custodian of gender-specific social norms. For the purpose of this study, two stage regression analysis characterized by the equations below has been used:
As before, Yi is the indicator for employment, Xi represents whether female ‘i’ lives in a joint family and Ci is the vector for individual and household controls. Fi (Mi) is the IV for joint family representing if the FIL (MIL) of the ith female is still alive. δi and φi are the error terms. Like before, state and district fixed effects are controlled and four variations of Yi have been used. The weak IV condition has been tested using Cragg-Donald Wald test. The null hypothesis in this test is that the IV is weak. If the calculated F statistic is within the ballpark range of 10, then we don’t reflect the null hypothesis. For both the IVs used here, the calculated F statistic was much over 10, and hence we can safely say that both the IVs do not suffer from the problem of weak IV. The regression results from the FIL still alive IV are reported in table 2, and that from the MIL still alive IV are reported in table 3.
6. Results
The results from IV regression using FIL still alive as an IV are significant at 1% (except for casual employment). If the likelihood of residing in a joint family goes up by 1 unit, then, ceteris paribus, on average, the likelihood of the female being employed in any kind of work reduces by 0.148 units. Column 2 (3) of table 2, depicts a similar negative relationship between self employment (salaried employment) and joint family. Joint family, however has an insignificant impact on employment of females in the casual sector. This maybe because of the fact that casual employment is often need based. When there is no or little income in the household, there is no option other than to take up any kind of available work irrespective of your gender. Although the direction of these estimates is same as that of the OLS estimates, they are higher in magnitude. This suggests a downward bias in the OLS estimates which could be attributed to the endogeneity problem. The results also suggest a positive U shaped relation between age and employment. This seems likely because middle aged women are more likely to work than maidens and elderly. The puzzling negative U shaped relationship between education and employment as observed by Sathar and Desai (2000) & Das and Desai (2003) is also evident here. This relationship is particularly puzzling because as the female acquires greater education, her skill set expands, which should increase her employability. But, the data suggests otherwise. There are various studies (Chatterjee, Desai and Vanneman (2019); Afridi, Mukhopadhyay and Sahoo (2016)) which propose factors like family income, spouse education and working hours as possible explanations. The availability of household help (maid) also has a small but positive and significant correlation with FLFP. This correlation seems obvious as the availability of a househelp, to some extent, frees the female from the burden of household duties like cooking and cleaning. The study also finds that Muslim, Sikh and other minority female are more likely to be employed than Hindu females. Residing in urban areas also has a negative correlation with employability of females. This might be because urban females generally belong to affluent families who care about their perceived social status. Allowing females to work outside the household harms this social status (Eswaran, Ramaswami and Wadhwa 2013).
Similar analysis was done using MIL still alive as an IV and coefficients of joint family on FLFP turned out to be insignificant (except for self employment). This suggests a greater role of patriarch in joint families. This result is in line with evidence from natural studies which highlight the social custom of living together while the patriarch is alive. Jain (2014) also highlights the greater role of patriarch in resolving conflicts among other members. Further, sons also have incentives for living under the same roof for as long as their father is alive because separation from the family might strip one’s rights on his father’s money and property.
7. Conclusion
According to IMF, the already low female employment rate in India (inclusive of formal and informal employment) of 35% in 2005, has gone down to 26% in 2018. The passive participation of females in the labor force is a huge hindrance for the economic development of the country. Currently, only 10 million out of 450 million women of working age are employed. According to World Bank, if all these eligible women joined the workforce, India could be 27% more richer. Moreover, as shown by glewee 1999 and Wodon 2013 employed females also play a significant role in raising living standards. Given these huge advantages of female workforce, the low FLFP rates are a matter of concern. The excessive pressure of household chores are often cited as reasons for resistance of females towards working outside the household. In this view, living in a joint family may seem as a possible solution. The members of a joint family can share the workload and resources, pool in time and income and extend a helping hand to the working female. This may reduce the burden of household chores and increase FLFP. But, this study provides contrary results. Using OLS and IV estimation, the study establishes a negative correlation between joint family and FLFP. The results are similar across different types of employment. As a measure of robustness check, other correlations are also in line with findings by other studies. Further, the study also suggests a greater role of patriarch in influencing the characteristic correlation between joint family and FLFP.
One of the limitations of this study concerns the choice of IV. FIL alive and MIL alive variables were not included in the 2005 dataset. One could conduct a panel data analysis by finding a better IV that is included in both the waves of the IHDS data. It would be interesting to see how effect of joint family on FLFP evolves over time. An important stage of economic development as observed worldwide is movement out of agriculture into the manufacturing and services sector. In that regard, it would also be interesting to segregate farm work and non-farm work and then analyze the impact of joint family, if any.
8. References
Afridi, F., Mukhopadhyay, A., Sahoo, S. (2016). Female labor force participation and child education in India: evidence from the National Rural Employment Guarantee Scheme. IZA Journal of Labor Development, 5(1), 7.
Carswell, G. (2016). Struggles over work take place at home: Women’s decisions, choices and constraints in the Tiruppur textile industry, India. Geoforum, 77, 134–145.
Chatterjee, E., Desai, S., Vanneman, R. (2018). INDIAN PARADOX: RISING EDUCATION, DECLINING WOMENS’ EMPLOYMENT. Demographic Research, 38, 855.
Chaudhuri, S., Morash, M., Yingling, J. (2014). Marriage migration, patriarchal bargains, and wife abuse: A study of South Asian women. Violence against women, 20(2), 141–161.
Das, M., Desai, S. (2003). Why are educated women less likely to be employed in India?: Testing competing hypotheses. Washington, DC: Social Protection, World Bank.
Debnath, S. (2015). The impact of household structure on female autonomy in developing countries. The Journal of Development Studies, 51(5), 485–502.
Desai, Sonalde, Reeve Vanneman and National Council of Applied Economic Research, New Delhi. India Human Development Survey-II (IHDS-II), 2011–12. ICPSR36151-v2. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2015–07–31.
Dhanaraj, S., Mahambare, V. (2019). Family structure, education and women’s employment in rural India. World Development, 115, 17–29.
Fernández, R. (2013). Cultural change as learning: The evolution of female labor force participation over a century. American Economic Review, 103(1), 472–500.
Greenwood, Jeremy, Nezih Guner, Georgi Kocharkov, and Cezar Santos. 2016. Technology and the Changing Family: A Unified Model of Marriage, Divorce, Educational Attainment, and Married Female Labor-Force Participation. American Economic Journal: Macroeconomics, 8 (1): 1–41.
Jain, M. (2016). Public pre-schooling and maternal labour force participation in rural India, Oxford Development Studies, 44(2), 246–263.
Jain, T. (2014). Where there is a will: Fertility behavior and sex bias in large families. Journal of Human Resources, 49(2), 393–423.
Khanna, M., Pandey, D. (2020). Reinforcing Gender Norms or Easing Housework Burdens? The Role of Mothersin-Law in Determining Women’s Labor Force Participation.
Maurer-Fazio, M., Connelly, R., Chen, L., Tang, L. (2011). Childcare, eldercare, and labor force participation of married women in urban China, 1982–2000. Journal of Human Resources, 46(2), 261–294.
Mukesh Eswaran Bharat Ramaswami Wilima Wadhwa, 2013. Status, Caste, and the Time Allocation of Women in Rural India, Economic Development and Cultural Change, University of Chicago Press, vol. 61(2), pages 311–333.
Sasaki, M. (2002). The causal effect of family structure on labor force participation among Japanese married women. Journal of Human Resources, 429–440.
Thomas, J. J. (2012). India’s labour market during the 2000s: Surveying the changes. Economic and Political Weekly, 39–51.
