Abstract:

The developing world is urbanizing very rapidly; while the average poor person was once a rural resident, today the average poor person lives in a city. The majority of urban growth is taking place in the poorest segments of urban society, both because of migration and high fertility among the urban poor. This has resulted in the growth and proliferation of slums in the developing world. This dissertation seeks to quantify what disadvantages are faced by children living in slums in India. As a large, quickly urbanizing, and democratic country, India is a natural place to undertake a study of slums.

I begin by examining how to best conceptualize and define slums, reviewing slum definitions from India and around the world. Following this, I use data from the National Family Health Survey 2005-2006, a national household survey of India that includes oversamples of slum populations, to examine three sets of outcomes. First, I consider infant and child mortality. I find that slum infants and children face higher mortality than other urban children but lower mortality than rural children. Once family background is taken into account, however, these differences disappear entirely. Second, I examine children's health. I find that slum children are no more likely to be malnourished or suffer acute illness symptoms than other children. Questions are raised about the validity of maternal reports of children's health, however. Third and last, I consider children's school attendance and work. Contrary to popular belief, I find that rural children are the most likely to be attending school net of family background characteristics. Slum children are indistinguishable from other urban children until age 14, at which point they become much less likely to be in school than either rural or other urban children. Though further research is needed, I attribute this to a combination of a vibrant informal economy in slums that provides incentives for children to work and a deficit of government run high schools for children to attend.

Overall, this dissertation provides some of the first rigorous quantitative analysis of the situation of slum residents in the developing world. Taken together, the results indicate that slums are a difficult and complex phenomenon to measure, not a homogenous set of neighborhoods. In some ways slums disadvantage residents, but in other ways slums may provide advantages compared to rural residence.