Assessing the Effect of Foster-Children Supply on Biological Children Education Demand : Some Evidence from Cameroon

In Cameroun, around 18 percent of children aged between 10-14 years old grow up within a sibship extended to host one (or more) foster-child. This proportion is similar in other African countries and in particular West African ones. This paper aims at estimating the effect of foster-children supply on biological children education demand of host parents in Cameroon. To address the endogeneity of foster-children supply, we estimate both decision within a recursive bivariate probit framework and use, as our identifying variable, the father's birth order among his brothers. Indeed, in patrilineal societies as in Cameroon, kinship rules involve children to be hosted by brothers of the male kin group, and more likely by the eldest. Using data from the demographic and health survey of Cameroon (2004), a dataset uniquely suitable for our purpose since information on the father's birth order are available, we find that children hosting school-age foster-relatives have a significant lower probability to attain their basic level of education relative to those who do not. This suggests that households hosting school-age foster-relatives due to kinship rules suffer from liquidity constraints preventing them from educating further their biological children. Through this result, we highlight the importance of the motive underlying child fostering to determine its spillover effects.