Intra-household allocation of educational expenses: Gender discrimination and investing in the future
Introduction
Discrimination against women, and in particular daughters, exists in many parts of the developing world. This is especially true in India, where a strong preference for sons has resulted in a skewed sex ratio due to female infanticide, sex selective abortions (Arnold, Kishor, & Roy, 2002), and the lack of proper diet and medical care for girls (Das Gupta, 1987). Sen (1990) notes that North Africa, China and South Asia have such severely skewed sex ratios that over a 100 million women are ‘missing' due to various forms of neglect.
Once born, girls face discrimination in the allocation of household resources. From the nutritional standpoint, Sen and Sengupta (1983) and a sharp bias against girls in terms of caloric intake in West Bengal and Behrman and Deolalikar (1990) report that the nutritional bur-den of a price rise falls disproportionately on women and girls. Jayachandran and Kuziemko (2011) and that mothers breastfeed daughters less than sons. A similar pattern has been noted in health expenditures, particularly when parents face a binding income constraint (Asfaw et al., 2010, Rose, 1999). Finally, a number of studies have found evidence of gender discrimination in educational expenditures in India (Azam and Kingdon, 2013, Kingdon, 2005, Lancaster et al., 2008, Subramanian and Deaton, 1991).
There are multiple social and economic factors that contribute towards a pro-male bias, particularly in educational expenditure. Aside from cultural preferences and social norms in favor of sons, there are differences by gender in expectations of old age support, perceived returns to schooling, costs of education and availability of schools (see Alderman and King (1998) for a summary and review of evidence supporting different motives for gender discrimination.) One channel through which the male bias may intensify is parents choosing to invest disproportionately in the child who is designated to look after them in their old age. Ebenstein (2013) proposes patrilocality (i.e. co-residence with sons) as a key determinant of the sex ratio in developing countries. In Korea, where patrilocality is the norm, he finds evidence that the sex ratio improves following a pension expansion which makes parents less likely to be dependent on their children in the future. While parents have higher expectations from sons in general, in India it is the eldest son in particular who is the family heir and typically assumes responsibility for his parents and extended family (Das Gupta, 1987, Mullatti, 1995). In Hindu families, the eldest son is also important for religious reasons and is responsible for performing the last rites of his parents. Jayachandran and Pande (2013) find evidence of better anthropometric outcomes for the eldest son in Indian families and attribute this to the special position that an eldest son enjoys.
This paper contributes to the literature on discrimination in education by using child specific data on educational expenditures and enrollment from the nationally representative India Human Development Survey-II (2011–12). The model uses household fixed effects controlling for age and birth order of the child. This methodology allows for an examination of intra-household patterns of discrimination. While previous literature has examined male bias in education, this paper additionally looks at differences in the treatment of not just sons, but eldest sons in particular.
I confirm the presence of a male bias and an additional preference for the eldest son in educational expenditures and enrollment. I also find that first born children receive preferential treatment. Next, I examine the behavior of households as the number of children increases and find that an increase in family size and greater competition for resources within the household causes the pro-male bias to fall and the bias in favor of the eldest son to become greater. As income increases and parents are less likely to depend on their children for support, preferential treatment towards the eldest son declines, particularly in the enrollment decision. Pro-male bias is strongest in the north, central and eastern states of the country. Finally, I study households in the state of Meghalaya, which follow a matrilineal system. In Meghalaya, as opposed to the rest of India, husbands move into the family home of their wives and the youngest daughter is the family heir (Gneezy, Leonard, & List, 2009). In line with the motivation to invest in the likely care giver, I find evidence suggesting reverse discrimination, i.e. discrimination against sons in this state.
The paper is organized as follows: Section two discusses motives for gender discrimination and makes the case for the importance of kinship norms. The data and sample are described in section three and section four discusses methodology. Results are presented in section five. The final section concludes and describes avenue for future work.
Section snippets
Gender discrimination and kinship norms
Kinship norms and social expectations play an important role in gender discrimination. In India, parents depend almost solely on their sons (and in particular the eldest son) for old age support and daughters are not expected to contribute to the material wellbeing of their natal families. Thus the perceived returns to educating a daughter are much lower than those for a son (Das Gupta, 1987, Foster and Rosenzweig, 1999). The prevalence of dowry in India makes daughters an additional liability
Data and sample
The India Human Development Survey 2011–12 (IHDS-II), which is jointly conducted by the National Council of Applied Economic Research in Delhi and the University of Maryland (Desai & Vanneman, 2011), is nationally representative and covers 42,152 households in 1503 villages and 971 urban regions across the country. The IHDS is a multi-topic survey that collects information on economic status, employment, education, demographics, health, gender relations and social capital.
In addition to
Methodology
I estimate the following household fixed effects model for each child in the sample:where Yih represents the outcome variable for child i in household h.μh are household fixed effects, EldestSonih is an indicator for being the first born male child and likely care giver, Maleih is the indicator for being a son, Ageih are indicators for the ages 6–18, and Firstih is an indicator for being the first child born in the family. Standard errors are
Educational expenses and the enrollment decision
Table 4 presents results for the main regression. Panel A presents results from the linear probability model for the enrollment decision. The sample comprises all children of school going age. Column 1 estimates the regression with household fixed effects, an indicator for being male, and age indicators, while Column 2 adds the indicators for first born, and eldest male child. The male bias is strong and persistent across both specifications. From Column 2 it is clear that parents give
Conclusion
This paper examines patterns of bias within the household in the allocation of resources to education. The analysis suggest that parents, who expect to live with and receive support from their eldest son in the future, invest the most in his education. The pro-eldest son bias is larger in larger families and those with lower incomes. In Meghalaya, which follows a matrilineal system, there appears to be some discrimination in favor of girls.
Gender discrimination within the household has broader
Conflict of interest
None.
Acknowledgements
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
References (34)
- et al.
Gender differences in parental investment in education
Structural Change and Economic Dynamics
(1998) - et al.
Are girls the fairer sex in India? Revisiting intra-household allocation of education expenditure
World Development
(2013) - Alderman, H., Orazem, P., & Paterno, E. (1996). School quality, school cost, and the public/private school choices of...
- et al.
Son preference, the family-building process and child mortality in India
Population Studies
(1998) - et al.
Sex selective abortions in India
Population and Development Review
(2002) - et al.
Gender gap in parents' financing strategy for hospitalization of their children: Evidence from India
Health Economics
(2010) - et al.
The intrahousehold demand for nutrients in rural south India: Individual estimates, fixed effects, and permanent income
Journal of Human Resources
(1990) - et al.
Birth order, schooling, and earnings
Journal of Labor Economics
(1986) - Coffey, D., Khera, R., & Spears, D. (2013). Women's health and children's height in India: Evidence from joint rural...
Selective discrimination against female children in rural Punjab, India
Population and Development Review
(1987)