Hunger in India: Facts and challenges — 3  

  Vol II : issue 6

  Amartya Sen
  Peter Svedberg
  M.S. Swaminathan
  Swadesh Deepak
Jayanta Mahapatra
  A.K. Shiva Kumar

  Only in Print

Subscribe to The Little Magazine
Order the print edition of this issue
Browse our bookstore
Browse back issues

   Mail this page link
   Enter recipient's e-mail:

Peter Svedberg

Gender Differentials in Anthropometric Failure Rates

4The evidence

All the seven surveys consulted here report separate data on male and female children. This data set is hence by far larger than those used in previous attempts to analyse differences along gender lines in India (for Africa, see Svedberg 1990; 1996). From this large data set, we have derived the ratios of the percentage of male and female children (M/F) who fall below the various anthropometric norms. The ensuing M/F ratios contradict the conventional view that females are at a systematic disadvantage. Had the conventional picture predominated, we would expect the M/F ratios to be consistently below 1.0. The M/F ratios for the four indicators and seven surveys (28 observations in all) are found to be below and above unity with equal frequency (Svedberg 2001b).

The M/F estimates presented in Table 4 are from the three survey periods for which data at the level of individual states are reported. With the weight for height indicator, the M/F ratio for all the states combined is above unity in all three periods and also in most of the individual states. There is no state where the ratio is below unity in all three surveys. With the two height-for-age indicators, the M/F ratios for all states pooled is slightly above unity in the 1974-79 and 1996-97 surveys, but below unity in 1988-90. At the state level, there is notable non-systematic variation. The M/F ratio for given states are above or below unity in different surveys and there is large variation across the states at given survey periods. Also, when it comes to underweight (W/A), no systematic pattern is revealed by the data.


In the absence of a more consistent pattern over time, and for lack of statistical significance, the tentative conclusion is that there is no systematic difference between boys and girls when it comes to low anthropometric status in India. But this conclusion is preliminary and has to be qualified on several accounts. First, only two surveys are national; the rest cover a limited number of states that may not be representative. Second, at the state level, the samples are uncomfortably small in many instances and some of the surveys are restricted to rural areas. Third, no investigation has been made of the M/R ratio in urban and rural areas separately. Fourth, the way the WHO has "transformed" the data from the Indian NNMB has yet to be scrutinised.


4Missing women?

There is also the question whether the non-systematic M/F ratios reflect more or less equal treatment of female and male children in terms of nutrition and healthcare — on the whole — or is a consequence of the under-representation of female children in the data. In most of the surveys consulted, there is a statistically significant difference in the number of male and female children observed. This under-representation may be the consequence of what Amartya Sen (1992, 2001) has labelled the "missing women" syndrome.

In a child population living under well-off and non-discriminatory conditions, the expected male to female ratio is about 1.05. The latest census for India (2001) gives a ratio of 1.07, down from 1.08 in 1991. In the seven anthropometric surveys consulted here, the M/F ratio in the sample sizes ranges from 1.04 in one survey to 1.11 in three surveys. The latter number indicates many "missing" female children. It may hence be that while there is little difference in the anthropometric status of male and female children in general in India, this can in part be a consequence of higher mortality rates for female infants and young children in the wake of discriminatory treatment. There are also mounting indications that the skewed child male/female ratio is a consequence of increasing gender differences in natality, i.e. sex-selective abortion (Sen 2001).


The present nutrition situation in India looks different depending on what measure of undernutrition is consulted. Estimates based on the national availability of food and its distribution across households from the FAO suggest that about one-quarter of Indian households have a habitual calorie intake that is insufficient to maintain health-consistent body weights and fuel light physical activity for the members. Estimates of the share of young children and adult women who are unduly short and underweight indicate that the prevalence of undernutrition is higher than purported by the FAO. The latter’s estimates are, however, unreliable for a number of reasons and should be interpreted very cautiously (Svedberg 1999; 2000; 2001a).

The share of extremely stunted children in India
has dropped from about 50 to 23 per cent.
Another encouraging finding is that the "conventional" notion that female children are at a systematic disadvantage vis-à-vis male children is not supported by the anthropometric data

The anthropometric indicators, although incomplete (covering young children and women only) and capturing illness as well as primary undernutrition, are more reliable markers of the nutrition situation. According to these indicators, the prevalence of undernutrition in India is (still) considerably higher than in the average African country and also slightly above the South Asian average (Appendix Table 1).

Looking at developments over time, there are some encouraging trends. The shares of young children who are abnormally short and underweight for their age have declined considerably since the 1970s. Especially notable is the fact that the share of extremely stunted children in India has dropped from about 50 to 23 per cent. Another encouraging finding is that the "conventional" notion that female children are at a systematic disadvantage vis-à-vis male children is not supported by the anthropometric data.

There are also discouraging developments. One is that the shares of children with unduly low height and weight for their age, while having declined more or less consistently for more than two decades, have increased in the late 1990s. It is too early to say whether this is an incidental statistical phenomenon or a true break in the previous trend; it is nevertheless worrying. Also discouraging is that inter-state differences, when it comes to height- and weight-for-age failure of children, seem to be growing. That is, while the developments for India as a whole look favourable, there are states where progress has been considerably slower. We only have estimates for eight states over time, but these indicate a growing diversification within India. Yet another discouraging development is that the urban/rural divide has grown in the 1990s.

While it is encouraging to find little systematic differences between male and female children in terms of anthropometric status, there are indications that this may be a consequence of a higher mortality rate for female infants and children. To find out more about this worrying link is one of the important challenges for future research, but there are others. In particular, no attempt has been made in this article to provide theoretical and empirical explanations for the statistical regularities revealed. To do that should also stand high on the research agenda for the near future.

p. 1 p. 2 p. 3 Sources

Peter Svedberg is Professor of Development Economics at the Institute for International Economic Studies, Stockholm University. He is the author of Poverty and Undernutrition: Theory, Measurement and Policy (Clarendon Press, Oxford, 2000, to be published by OUP India soon), a definitive study of undernutrition in Sub-Saharan Africa and South Asia. He lives in Stockholm