This paper tests for the existence of a Poverty Nutrition Trap (PNT) in the case of the nutrient most likely to have productivity impacts, i.e., calories, for three categories of wages - sowing, harvesting, and 'other' - and for male and female workers separately. We use household level national data for rural India for the period January to June 1994. We use robust sample selection procedures based on Tobit and Heckman methods to arrive at consistent estimates. It was found that a PNT exists for women workers engaged in harvesting and sowing in the case of the Heckman methodology. In the case of the Tobit analysis a PNT exists in the case of female harvest, male 'other', and female 'other' categories of wages.
ASARC Working Paper 2006-2, 24 pp.