Human Development

Gender composition of children and sanitation behaviour

  • Blog Post Date 22 April, 2020
  • Print Page
Author Image

Deepak Saraswat

The University of Connecticut

Open defecation has been linked to various public health issues and has gained significant policy attention. Nonetheless, research has shown that due to son preference, households in India underinvest in outcomes for their female children. Using the 2015-16 Demographic and Health Survey, this article finds that households reduce open defecation in the presence of female children, as they face a potentially high cost of harassment related to them.


India accounts for 60% of the world’s open defecation (OD), and OD is a major source of poor sanitation in India. Poor sanitation impacts public health through substantial negative externalities. While improvements in sanitation have been linked to improved health and human capital in developed countries such as the US (see Bleakley 2007, Cutler and Miller 2005, Watson 2006), low levels of sanitation have been linked to acute malnutrition, diarrhoea, anaemia, and many other health problems stemming from faecal pathogens in developing countries (see Coffey et al. 2016). Sanitation and reduction in OD has been a focal point for policymakers and academic research in public health and economics. When it comes to the need for reduction in OD, the stakes are quite high; Geruso and Spears (2018) estimate that a 10-percentage-point reduction in neighbourhood OD is associated with a 10% reduction in infant mortality (by 6.1 children per 1,000).

Sanitation and women in India: Why do we care?

Women value toilets to a greater extent than men as they require much more privacy and hygiene when defecating in the open or attending to menstrual hygiene. Women are also disproportionately subjected to harassment when defecating in the open1. India, like many other developing countries, has a widespread preference for sons over daughters, and this skewed preference has led to selective under-investment in female children2. A natural question then comes up: why would then a household care to invest in a sanitation facility for a girl child? Digging deeper, we may find an answer in how Indian society is organised. While any private benefits (such as education, nutrition, etc.) to a female child can be curtailed since doing so only imposes a private cost on her, the cost of OD can manifest in the form of associated harassment for the child, which imposes a cost on the household as a whole. Religious doctrines in India assert that women are to be dependent on the men in the family and are their responsibility, hence any harassment associated with them can be linked to the entire household, in terms of shame. I subject this hypothesis to an empirical test and find evidence in favour of it.


Main results

Using the 2015-16 DHS (Demographic and Health Survey) in India and following the empirical methodology in related literature (Barcellos et al. 2014, Bhalotra and Cochrane 2010, Jayachandran and Kuziemko 2011), I use the gender of the firstborn child being female as an indicator of the presence of female children in Indian households, and find that it leads to a reduction in OD amounting to about 7% in rural and 14% in urban areas3.

Differential response in rural and urban areas

Rural and urban regions differ in their costs from and incentives to reduce OD. In comparison to urban areas, rural areas have large spaces and hence better avenues to get privacy when defecating in the open. Urban areas have much higher crime against women than their rural counterparts4. Defecating in open also has much higher social costs (in terms of shame from spreading impurity in the neighbourhood) in urban areas as compared to rural areas. My calculation using DHS shows that while OD demonstrates a sharp reduction in urban areas as income goes up, the reduction in rural areas is much less with income going up. Therefore, one must expect that in urban areas, the households who are likely to be compliers are the ones that have high rates of OD and bear the most costs from it, which is likely to be the poor pockets. On the other hand, in rural areas, the income constraint to invest in a toilet is higher with less costs from OD and hence, the compliers are likely to be the richer households. Results suggest that this indeed has empirical backing. The reduction in OD due to firstborn child being female is more pronounced in poor urban households and in richer rural households.

Reduction in OD comes up when it matters the most

Results suggest that the reduction in OD due to firstborn child being female shows up only when the girl child has reached the age of puberty (see Khadgawat et al. 2016), and not when the firstborn child is younger. Further analysis suggests that in both urban and rural areas, the reduction in OD is more pronounced when the firstborn child is closer to the age of marriage of girls in both regions. I also find evidence that the reduction in OD from main results are largely driven by areas where the crime against women is the highest in the country (using IHDS, 2011-12). States with low socioeconomic and development indices5 are also likely to be the ones where State capacity is weaker in protecting girls from harassment, and the results are disproportionately stronger in these regions. These variations suggest that households are indeed investing in reducing OD when it is imperative to protect their girl child from any potential harassment and when dependence on State capacity for protection is weaker.

What do we learn?

Sanitation has been looked at through the lens of a ‘public good’ problem in economic literature where, due to substantial negative externalities, private investment in the reduction of OD is suboptimal. This research highlights that in certain cases, there are private costs from OD and hence, private investment is possible. This research also contributes to the economic literature on skewed gender preferences for children and the absence of private investment in girls, and highlights that in the presence of costs that are non-exclusive to girls, households have incentives to invest in private benefits for a girl. Association between the gender of the firstborn child and sanitation practices of a household in India also provides a new first-stage result, which has the potential to generate higher take-ups with targeted incentives to families with presence of girl child. Finally, these findings are relevant for policymakers concerned with improving the targeting of resources for sanitation in India.


  1. They are twice more likely to suffer from non-partner sexual violence when defecating in open (Jadhav al. 2016). In urban slums of New Delhi, about 66% women report verbal abuse and 50% report visual abuse (JAGORI and UN Women, 2010).
  2. Less investment in education, nutrition, and postnatal time and attention (for example, breastfeeding) (Barcellos et al. 2012, Jayachandran and Kuziemko 2011, and Deaton 2003).
  3. This result holds up to multiple stress tests in form of robustness to various geographic controls, set of fixed effects, and their interactions.
  4. Author’s calculations using Indian Human Development Survey (IHDS), 2011-12.
  5. BIMARU states (Bihar, Madhya Pradesh, Rajasthan, and Uttar Pradesh) and states with low levels of literacy.

Further Reading

  • Barcellos, Silvia H, Leonardo S Carvalho and Adriana Lleras-Muney (2014), “Child Gender and Parental Investments in India: Are Boys and Girls Treated Differently?”, American Economic Journal: Applied Economics, 6:157-189.
  • Bhalotra, S and CT Cochrane (2010), ‘Where Have All the Young Girls Gone? Identification of Sex Selection in India’, IZA Discussion Paper no. 5381, Institute for the Study of Labor, Bonn.
  • Bleakley, Hoyt (2007), “Disease and Development: Evidence from Hookworm Eradication in the American South”, Quarterly Journal of Economics, 122(1):73-117.
  • Coffey, Diane, Michael Geruso and Dean Spears (2017), “Sanitation, Disease Externalities, and Anemia: Evidence From Nepal”, The Economic Journal, Royal Economic Society, 128(611):1395-1432.
  • Cutler, David M and Grant Miller (2005), “The role of public health improvements in health advances: The twentieth-century United States”, Demography, 42(1):1-22.
  • Deaton, Angus (2003), “Health, Inequality, and Economic Development”, Journal of Economic Literature, 41(1):113-158.
  • Geruso, Michael and Dean Spears (2018), “Neighborhood Sanitation and Infant Mortality”, American Economic Journal: Applied Economics, 10:125-162.
  • Jadhav, Apoorva, Abigail Weitzman, and Emily Smith-Greenway (2016), “Household sanitation facilities and women’s risk of non-partner sexual violence in India”, BMC Public Health, 16(1139).
  • JAGORI and UN-Women (2010), ‘Safe Cities Free of Violence Against Women and Girls Initiative: Baseline Survey in Delhi’, Report.
  • Jayachandran, Seema and Ilyana Kuziemko (2011), “Why Do Mothers Breastfeed Girls Less than Boys? Evidence and Implications for Child Health in India”, Quarterly Journal of Economics, 126:1485-1538.
  • Khadgawat, Rajesh, RK Marwaha, Neena Mehan, Vineet Surana, Asshima Dabas, V Sreenivas, M Ashraf Gaine and Nandita Gupta (2016), “Age of Onset of Puberty in Apparently Healthy School Girls from Northern India”, Indian Pediatrics, 53(5):383-387.
  • Watson, Tara (2006), “Public health investments and the infant mortality gap: Evidence from federal sanitation interventions on U.S. Indian reservations”, Journal of Public Economics, 90(8-9):1537-1560.
No comments yet
Join the conversation
Captcha Captcha Reload

Comments will be held for moderation. Your contact information will not be made public.

Related content

Sign up to our newsletter