Breaking the Net
Family Structure and Street Children in Zambia
Published In: Brooks World Poverty Institute Working Paper, no. 111 (Manchester, UK : Brooks World Poverty Institute, Jan. 2010), 26 p
The safety net provided by the African extended family has traditionally been the basis for the assertion that "there is no such thing as an orphan in Africa" (Foster 2000). The assumption is that even families lacking sufficient resources to properly care for existing members are predisposed to take in orphans. Chronic poverty, coupled with an increasing malaria burden and the HIV/AIDS pandemic, has put this safety-net under severe strain, giving rise to an increasing number of orphans and vulnerable children and, in the extreme, to "street children." Drawing on original fieldwork in the slums of Ndola in Northern Zambia we study the role of family structure in caring for vulnerable children. We try to isolate those features of a child's nuclear and extended family that put him most at risk of ending up on the streets. We find that older, male children and particularly orphaned children are more likely to wind up on the street. Families with a male household head who is in poor health are more likely to originate street children. The educational level, age and employment status of the male head of household has little impact on the likelihood the family is associated with a child who has taken to the street. In contrast, households with surviving maternal grandparents or with a male head who has many sisters are significantly less likely to originate street children. These findings support the critical role that women play in poor countries, highlighting the importance of policies aimed at empowering women. At the same time, our findings show that policies aimed at improving the health of the male head of household can also yield important benefits. A back-of-the-envelope calculation suggests that moving male heads from poor to good self-rated health status can increase the rate of GDP growth by as much as 0.20 to 0.33 of a percentage point per year.