Child Malnutrition in Shining India:Peter Svedberg
|
The core question addressed in this paper is what explains the large inter-state differences in child malnutrition as manifested in stunting in India⎯ranging from 22 to 56 per cent. The method used is multiple regression analysis with controls for multicolinearity, reverse causation and robustness.
The major findings in this paper are two-fold. The first is that the variance in prevalence of child stunting across the Indian states is strongly correlated to the health care provided to children. The second is that child health-care provision is mainly determined by household per-capita income and the relative status of women. Both results are robust to the use of alternative measures of health-care provision and female status and are unlikely to reflect reverse causality. The health-care variable is not only highly significant; the correlation coefficient suggests that a higher coverage of child health care has a substantial quantitative effect on the incidence of child stunting (also see Figure 1) A simple simulation exercise suggests that if the child health-care provision in the seven states where it is presently below the median, was brought up to this level (37.5 per cent), the prevalence of child stunting in all-India would be reduced by about 9 per cent (or by 3.9 percentage points). This would not bring an end to child malnutrition, but considering that the prevalence of child stunting in India remained unaltered over the 1990s [Svedberg 2006], a decline of magnitude is not negligible. From a broader policy perspective, the findings in this paper indicate that there are no simple remedies for alleviating child malnutrition in India over the short term. The many intertwined links through which child stunting apparently is affected, means that no single policy measure can be expected to be the panacea. That high economic growth is not sufficient (although pro-poor growth is most likely necessary) for reducing child malnutrition is suggested by the insignificant correlation between child stunting and levels of per-capita income across the states. The overall high growth of per-capita NDP 27 over the 1990s in India was paired with deteriorating income inequality in all dimensions, the inter-state, the rural-urban, the within rural, the within urban, and across households16 ⎯ most probably the main reason why child stunting did not decline. The findings reported here suggest that a complex combination of policy interventions is required, focused on income growth that involves the poor population segments, female education, fertility reduction and ⎯ as a more short-term policy ⎯ extended provision of high-quality health-care for young children. That the federal Indian government has limited juridical and financial power over health-care provision in the states is an obstacle for concerted action at this level, but an overhaul of the inefficient federally financed ICDS program [das Gupta et al. 2005] would be a welcome initiative. The entrenched cultural traits that lie behind the subjugation of females in large parts of rural and urban India can be expected to change only in the long term. * Peter Svedberg is professor of development economics at the Institute for International Economic Studies, Stockholm University. E-mail: peter.svedberg@iies.su.se. Read the full document ekduniya socialjustice |
Printable version
Back to top
