Confronting Amartya Sen's view: Why WHO's methodology on malnutrition is hopelessly flawed for India

In a recent TV interview, when the anchor asked Amartya Sen how he would respond to economists Arvind Panagariya andSwaminathan Aiyar who had questioned the basis of his estimate of 1,000 deaths per week due to non-implementation of the food security Bill, he prefaced thus, "Panagariya I don't think actually believes there is muchundernourishment in the country. He thinks this is a myth - at least this is the headline of his paper."

The remark bore no direct relationship to the question posed and was perhaps intended to discredit its source, namely, me. It reminded me of my only encounter with Sen in a TV debate on the food security Bill in which he disparaged me by saying that being resident in New York I wouldn't understand the ground reality of India. Ironically, it didn't dawn on him that he had spent an even larger proportion of his life abroad!

FLAWS IN THE METHOD

Sen's latest undernourishment remark fares no better. It reveals that rather than read my paper he had inferred its content from the title, which, as he knows well, is chosen by the newspaper editors in the first place. The remark also lays bare that his collaborator Jean Dreze, who had corresponded with me about the paper immediately after its publication, never discussed with him my objections to malnutrition estimates they have uncritically reported in their book. Under normal scholarly practice, you would record such objections even if you reject them.

For readers unfamiliar with the context of Sen's remark, let me explain. I have argued in several of my recent writings, including a detailed article in the Economic and Political Weekly, that the current one-size-fits-all methodology to estimate the incidence of malnutrition is hopelessly flawed. Aggressively pushed by the World Health Organization(WHO), this methodology applies a globally common set of height and weight norms forchildren of given age and gender to determine whether a child of that age and gender is stunted (low height for age) or underweight (low weight for age).

THE CONTEXT MATTERS...

Underlying this methodology is the assumption that different populations of children of given age and gender have identical height and weight potential regardless of race, ethnicity, culture and geography. In other words, with fully nutritious diet, on average, five-year old Indian boys will attain the same height and weight as Norwegian five-year old boys. But empirical evidence contradicts this assumption. Race, ethnicity, culture and geography do matter.

Thus, for example, a study shows that after controlling for maternal socio-demographic and prenatal care factors, on average, newborn children of Japanese-American mothers are smaller and weigh less than those of American mothers. Children of Indian-American mothers show the exact same pattern. Yet, the India-American children are as healthy in terms of infant mortality rates as their white-American counterparts and healthier than black-American counterparts. In a similar vein, five-year old girls of Moroccan descent in the Netherlands remain 2.7 centimeters shorter than their Dutch counterparts.