Inequality and child height

The Impact of Poverty and Discrimination on Child Height in India

Diane Coffey, Ashwini Deshpande, Jeffrey Hammer, and Dean Spears 

iframe Tweet

Introduction

Child height is an important marker of population-level health because it reflects children’s exposure to malnutrition and infectious disease. Inequality in child height in population subgroups indicates inequality in early life health more broadly, which leads to unequal adult outcomes. Adults in population subgroups whose early life experiences allow them to grow taller have better cognitive and economic outcomes, on average. Differences in child height, therefore, are a harbinger of continued inequality, especially in poor countries with high burdens of malnutrition and infectious disease, such as India. 

It is not surprising then that children from lower socioeconomic status (SES) families would be shorter than children in higher SES families – economic resources can buy important health resources such as a food and clean cooking fuel. But socioeconomic status does not always account for differences in height. Forms of discrimination, which are not measured in SES, could also affect child height. The authors ask: among children at the same level of SES, are those from social groups that suffer discrimination shorter? 

Using data on children in rural households from the 2005 National Family Health Survey-3, India’s Demographic and Health Survey, the authors study height differences among children from four population groups: Scheduled Tribes, Scheduled Castes, Other Backward Castes, and general castes (see Box). The authors calculate what the average height of Scheduled Tribe, Scheduled Caste, and Other Backward Class children would be if they had the same distribution of socioeconomic characteristics as general caste children.

The authors find that socioeconomic differences explain why Scheduled Tribe children, who often grow up geographically and socially isolated from other population groups, are shorter than general caste children. However, socioeconomic variables alone do not explain why Scheduled Caste children are shorter than general caste children. Instead, height gaps between Scheduled Caste and general caste children are explained by a combination of SES variables and the fraction of households in a Scheduled Caste child’s neighborhood that outrank her household in the caste system. This may be because pregnant mothers experience more stress when they live near general caste neighbors or it may be because of the particular ways in which the general castes enforce their social dominance over Scheduled Castes, such as limiting Scheduled Caste families’ use of clean water.

Box: Caste, Tribe, and Inequality in India

Scheduled Tribes: People from Scheduled Tribes belong to adivasi (indigenous) groups and have a social identify outside the caste system. Ninety percent of adivasis live in rural areas, often in isolated villages that are not well served by public resources. People from adivasi backgrounds are among the most economically and educationally deprived in India Schedule Tribes are eligible for affirmative action in education and government jobs.

Scheduled Castes: Scheduled Castes are those which once were, and in some circumstances still are, treated as “untouchable.” People from Scheduled Castes often live in the same villages as people from higher caste backgrounds. They often face discrimination in their interactions with people from higher castes. Scheduled Castes are eligible for affirmative action in education and government jobs.

Other Backward Classes: “Other Backward Classes” is a term used by the Indian government to describe people from castes or communities that were not considered untouchable, but have historically been disadvantaged relative to the general castes. They typically do not face the same stigma as people from Scheduled Castes. Other Backward Classes are eligible for affirmative action in education and government jobs.

General caste:  General castes are socially and economically advantaged compared with Scheduled Tribes, Scheduled Castes, and Other Backward Classes. They are ineligible for affirmative action programs.

Key Findings

• Scheduled Caste, Scheduled Tribe, and Other Backward Class children are shorter, on average, than general caste children.
• Socioeconomic status (SES) variables explain height gaps between Scheduled Tribe and general caste children, but cannot fully explain gaps between Scheduled Caste and general caste children, nor between Other Backward Classes and general caste children (see Figure 1).
• Among Scheduled Caste and Other Backward Class children, SES differences combined with local processes of discrimination matter (see Figure 2).
  • Specifically, Scheduled Caste and Other Backward Class children who live in neighborhoods with more households that outrank their own in the caste system are shorter than general caste children of the same SES.
  • On the other hand, Scheduled Caste and Other Backward Class children who are surrounded by other lower caste households are no shorter than higher caste children of the same SES.

Figure 1. Socioeconomic status differences fully explain why Scheduled Tribe children are shorter than general caste children, but not why Scheduled Class and Other Backward Class children are shorter  

Within each pair of results, which show the point estimates (dots) and 95% confidence intervals (bars), the results on the left display the difference in standard deviations of height-for-age between a socially disadvantaged group of children and general caste children. The results on the right display the difference in height after differences in *socioeconomic status (SES) are accounted for. SES variables explain the height gap for Scheduled Tribe children (the bar crosses zero), but do not fully explain the gaps with Scheduled Caste and Other Backward Caste children because height gaps still remain.

Figure 2. After taking into account socioeconomic status, the remaining height gap for Scheduled Caste and Other Backward Caste children is explained by the fraction of a child’s neighbors that outrank her family in the caste system

This figure shows that, after controlling for sex, age, and household wealth, Scheduled Caste and Other Backward Class children who live in neighborhoods where they are surrounded by a larger fraction of households that outrank their own family in the caste system are shorter than Scheduled Caste and Other Backward Class children who live in neighborhoods where they are surrounded by a smaller fraction of households that outrank their own family in the caste system.

Policy Implications

These results raise questions about the limits of affirmative action policies to address the effects of discrimination against socially disadvantaged groups in India. Though affirmative action policies have been successful in improving educational and employment outcomes for some members of lower castes, they are insufficient to respond to local discrimination whose impact on early health can have lasting effects on cognitive development and future prospects. Social policies must also target local processes of discrimination to weaken the link between social background and poor health.

Reference

Coffey, D., Deshpande, A., Hammer, J., & Spears, D. (2019). “Local social inequality, economic inequality, and disparities in child height in India,” Demography (Online ahead of print). 

Suggested Citation

Coffey, D., Deshpande, A., Hammer, J., & Spears, D. (2019). The impact of poverty and discrimination on child height in India. PRC Research Brief 4(2). DOI:10.26153/TSW/954. 

About the Authors

Diane Coffey (coffey@utexas.edu) is an assistant professor of sociology and a faculty research associate in the Population Research Center, The University of Texas at Austin; Ashwini Deshpande is a professor of economics at Ashoka University; Jeffrey Hammer is the Charles and Marie Robertson Visiting Professor in Economic Development at Princeton University; and Dean Spears is an assistant professor of economics and a faculty research associate in the Population Research Center, UT Austin.

Acknowledgements

This research was supported by grant P2CHD042849, awarded to the Population Research Center at The University of Texas at Austin by the Eunice Kennedy Shriver National Institute of Child Health and Human Development. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.