Birth Weight and Early Childhood Physical Health: Evidence from a Sample of Latin American Twins
In lieu of an abstract, here is a brief excerpt of the content:

Birth Weight and Early Childhood Physical Health:
Evidence from a Sample of Latin American Twins

Low birth weight is considered one of the leading causes of infant mortality and adverse health conditions during childhood and throughout life. Most developed countries have therefore implemented health policies related to the improvement of birth weight of newborns. Examples include Medicaid and the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC), both in the United States. Most of the existing literature evaluating the effects of nutritional programs targeted to pregnant and nursing women in developed countries emphasizes their positive effects on birth outcomes and the physical development of children, yet developing countries are still resistant to implementing social welfare programs aiming to enhance the nutritional status and prenatal care of pregnant women in order to improve the health of their babies.1

There is, however, an open discussion on whether birth weight is important in determining health status profiles and labor productivity later in adulthood or if this indicator of early nutritional status (arguably, in utero nutritional condition) captures other unobservable factors. Moreover, empirical evidence [End Page 161] is based on data sets linking clinical birth records with outcomes later in life, potentially introducing additional biases such as attrition and selection caused by the survival condition of individuals, reinforcing skepticism about the results. Altogether, these facts call into question the role of birth weight as a key indicator for the design of health-related policies and the effectiveness of government policies attempting to improve the healthcare of expectant mothers to reduce the prevalence of low birth weight.

This paper focuses on determining the causal effect of birth weight on physical health for a sample of children in developing countries and contributes to the empirical literature regarding the effects of birth weight throughout the life span in three different ways. First, in contrast to previous empirical studies focusing on adult outcomes, this paper examines the effects of birth weight on the physical development of children under age five. Early developmental stages can have long-term effects on health, educational, and labor market outcomes, which can mediate the birth weight and educational or labor market gradients during adulthood.2 Second, in contrast to the existing literature focusing solely on developed countries, the paper provides evidence for ten Latin American countries using data from Demographic and Health Surveys (DHS). Finally, information included in the data set allows testing whether postnatal health investments (namely, breastfeeding and vaccination) can potentiate health and nutritional investments made by parents during periods prior to the birth of their children.

One particular concern is that the effect of birth weight on the physical health of infants might be spurious due to omitted unobservable factors. In this regard, children’s birth weight and physical growth are mutually determined by genetics, environmental conditions to which the mother was exposed during pregnancy, family background, and socioeconomic status, among other factors. For this reason, cross-sectional estimates of the effects of birth weight on physical growth may contain confounding factors.

To isolate the effects of variations of birth weight from unobservable factors, I use twin-pair fixed effects from a sample of twins born in ten Latin American countries. Information on twins’ weight at birth and physical health is obtained from a pooled sample of different years from the DHS data sets of each country. Since twins are exposed to the same conditions during pregnancy and the very moment of birth (for example, maternal stress, length of pregnancy, and institutionalized delivery), twin-based estimates successfully [End Page 162] identify the effect of physical growth under age five. The key insight of this approach is to assume that differences in birth weight between twins are purely due to nutritional intake within the womb, which depends on either the fetal position or the connection between the umbilical cord and the placenta, both of which are determined by nature.

There are five important findings. First, consistent with previous empirical studies, cross-sectional estimates indeed overstate the effect of birth weight on children’s physical growth relative to twin-based estimates. Second, results arising from twin comparisons suggest that increasing birth weight significantly increases the height...