Academic journal article Demographic Research

Children's Stunting in Sub-Saharan Africa: Is There an Externality Effect of High Fertility?

Academic journal article Demographic Research

Children's Stunting in Sub-Saharan Africa: Is There an Externality Effect of High Fertility?

Article excerpt


A positive relationship between the number of siblings and a child's chance of being stunted has been seen in several studies. It is possible that individual stunting risks are also raised by high fertility in the community, partly because of the impact of aggregate fertility on the local economy, but this issue has not been addressed in earlier investigations. In this study we estimate the independent effect of the child dependency ratio in the province (or governorate, region, or larger geopolitical zone within a country), using DHS data on up to 145,000 children in 152 provinces in 23 countries with at least two such surveys. The data design allows inclusion of lagged province variables and province fixed effects (to control for constant unobserved province characteristics). Three types of regression models for a child's chance of being stunted are estimated. Some estimates suggest an adverse effect of the current child dependency ratio, net of the child's number of siblings, while others do not point in this direction. When the child dependency ratio measured in an earlier survey is included instead, no significant effects appear. Thus, we conclude that there is only weak support for the idea that a child's stunting risk may be raised by high fertility in the community.

(ProQuest: ... denotes formulae omitted.)

1. Introduction

Under-nutrition of children is unfortunately still very common in many parts of the world. For example, 42% of the children in sub-Saharan Africa are stunted (i.e., have a low height for their age) - an indicator of chronic under-nutrition (UNICEF 2010). In addition to the serious implications for child mortality, there are long-term effects of childhood under-nutrition on health, well-being, and productivity (Black, Morris, and Bryce 2003; Chang et al. 2002; Glewwe, Jacoby, and King 2001; Graff et al. 2010; Grantham-McGregor et al. 2007; Mendez and Adair 1999; Weinreb et al. 2002).

A child's nutritional status is to a large extent determined by his or her food intake and exposure to diseases, and the treatment for these, which are in turn influenced by a number of individual, household, and community factors. One household factor that has attracted much attention as a potential determinant of stunting is the number of siblings below certain ages. It has been concluded in several studies (see references below) that children with many siblings are particularly likely to suffer from under-nutrition - not least because of dilution of resources - although one can of course never be sure that joint determinants of fertility and nutritional status have been adequately controlled for. Such a 'sibling effect' may be considered part of a more general health and economic disadvantage of children in large families (e.g. Anh et al. 1998; Cleland et al. 2006; Eloundou-Enyegue and Williams 2006; Li, Zhang, and Zhu 2008) and their mothers (e.g. Montgomery and Lloyd 1996).

The objective of this study is to find out whether, in addition to the effect of many young siblings in the household, high fertility among other people in the community may increase a child's chance of being stunted. Such an effect has not been documented in the literature, but seems plausible for two main reasons. First, a larger number of children in other families may increase the chance of childhood diseases in these families, because of resource dilution and because a child may have a higher probability of getting an infectious disease when the household is more crowded. This higher prevalence of diseases in the community may increase the stunting risk for any particular child. Second, there may be aggregate-level economic effects of high fertility. Although earlier econometric studies have provided rather mixed results (Headey and Hodge 2009), there is some support for the idea that income growth is hampered by a large relative size of the young population, as measured by current birth rates (Kelley and Schmidt 1995) or the proportion of young (Kelley and Schmidt 2005). …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.