How are foster children treated relative to non-foster children? Do they consume less than other members of the household? Child fostering is ubiquitous in Africa and the wellbeing of these children, who may be particularly vulnerable to impoverishment, is not well known. But assessing poverty for individual children can be a daunting task, since consumption is typically measured at the household level and many goods in the household are shared. To address this problem, I extend Dunbar et al. (2013) and estimate a collective model of household behavior designed to capture resource shares, that is, the share of total household expenditure allocated to each household member. Resource shares are identified by observing how expenditure on assignable clothing varies with income and household size. I show that the data requirements for estimating individual poverty incidence are less demanding than previously thought under reasonable restrictions to how resource shares vary across household structure. I find that standard poverty indices substantially understate foster child poverty, and that foster children consume a slightly smaller share of household resources than nonfoster children. This difference is driven by orphaned foster children, suggesting kinship networks play an important role in foster child treatment.