You see, stunting rates in India are comparable to Malawi. This fact is not inconsistent with my observations during travels through India (mostly as a young backpacker and later short trips for work) and the large amounts of time I spent in Malawi. However, as a development economist, I don’t rely on my casual observations, but on data. And, the data tell me that India is richer and does better in a whole host of health and development outcomes. So, the fact that close to one in two children is stunted in both countries is simply jarring.
In a shorter paper  for India’s EPW, the authors take on Arvind Panagariya’s argument that the differences are genetic (see here  for an example of the argument made by Prof. Panagariya  for an India-specific metric rather than the WHO’s stunting measures) and systematically dismantle it. The authors’ first observation is that in both India and in 25 sub-Saharan African (SSA) countries in their data, there is birth order gradient in height-for-age z-score (HFA): it is highest for first-born children and declines for higher order births. However, the authors also observe that this gradient is much steeper in India: while the increase in stunting for the 2nd and the 3rd + child is about 4.5 and 9.3 percentage points (pp), respectively, compared to the first child in SSA, the same figures are 15 and 24 pp. So, while first-born Indian children grow up to be taller than children in SSA, the second-born fall behind and the third- and higher-order births fall substantially behind. This remains true when the authors analyze z-scores instead of a binary measure of stunting and use only within-family variation.
It’s quickly obvious that a genetic explanation is inconsistent with the pattern presented above. What else could be going on? The authors rule out the possibility that higher child survival in India means that we observe the shorted children there but not in SSA (infant survival needs to be higher for higher-order births in India for this to be a plausible explanation, where India actually has the advantage among the first-order births). Lower mother’s height in India is also unlikely to have differential impacts across birth order. So, the authors focus on an explanation that can be explained through the economics of the household: household resources are expanded less on the mother and her children through the birth order and the take-up of health services similarly declines.
For example, consumption declines among women across successive pregnancies, but not among their husbands: so Indian households may disinvest in the mother more across successive pregnancies. Child health may therefore suffer through impacts of the mother’s nutrition status during pregnancy and while breastfeeding. Take-up of prenatal check-ups, nutritional supplementation, child vaccinations, postnatal check-ups all decline faster in India than in SSA across birth order. But, why…
The authors think that the answer lies in the stronger son preference in India, especially the cultural norm of eldest son preference (a preference for first children is seen in most societies, while son preference is strong in South Asia). The authors describe the evidence best: “…once the family has a son in India, prenatal inputs decline with subsequent pregnancies. Girls born before the eldest son actually fare better than boys born after the eldest son, and girls born after the eldest son fare the worst. Consistent with this, the data show that at birth, the height differential between boys and girls is similar in India to Africa; however, over time significant gender gap emerges wherein the height differential between boys and girls in India exceeds that in Africa.”
The question is how to tackle the norms and conditions that lead to unequal resources available to women over the course of their marriages and the strange inequality of opportunity that arises across birth order arising from a more common inequality of opportunity between boys and girls.
I am assigning the paper to my students next year…