Duke 2008 | ASHE | Equity and Efficiency in Health and Healthcare

[I]ncreasing breastfeeding rates throughout the United States and?promoting optimal breastfeeding practices,' has been identified as an important public health goal by the Center for Disease Control (CDC, 2007). As part of Healthy People 2010 (2000), the U.S. Department of Health and Human Services has targeted a 75% breastfeeding participation rate during the early postpartum period. Breastfeeding rates have been on the rise since the early 1990s (CDC, 2007), and an increasing number of state and federal regulations have been enacted to protect the rights of mothers who breastfeed their infant children. One of the impetuses for promoting breastfeeding is the claim that breastfeeding reduces the incidence of childhood illnesses and chronic disease (American Academy of Pediatrics, 1997). However, it is also asserted that breastfeeding may yield important short- and long-run cognitive development benefits. While a number of studies in the medical, nutrition, and cognitive development literatures have found evidence of a positive association between breastfeeding and child intelligence (Mortensen et al., 2002; Anderson et al., 1999; Jacobson et al., 1999; Golding et al., 1997), there is a growing concern that these findings may be confounded by family-level unmeasured heterogeneity. In particular, a recent analysis of sibling pairs in the National Longitudinal Survey of Youth 1979 (NLSY79), found that breastfeeding had no effect on childhood Peabody Individual Achievement Test (PIAT) scores (Der et al., 2006). These results have been interpreted as evidence that breastfeeding has little or no effect on cognitive development or academic achievement. However, the robustness of this finding not yet been tested using other nationally representative datasets, nor have a wider set of long-run schooling outcomes been examined. The current study contributes to the existing body of literature by using a nationally representative dataset that has not been exploited in this literature?the National Longitudinal Study of Adolescent Health?to estimate the relationship between breastfeeding and four long-run measures of schooling: the Peabody Picture Vocabulary Test (PVT), cumulative high school grade point average, high school graduation, and college attendance. Family fixed effects estimates using data on brother pairs show no evidence of positive breastfeeding effects, and these estimates are precise enough to reject OLS estimates. However, for females, family fixed effects models show evidence of a positive relationship between breastfeeding and long-run academic performance. Controlling for time-invariant family-level heterogeneity, breastfeeding is associated with a 4.30 point increase in early adult PVT scores and a 10 percentage-point increase in the likelihood of high school graduation. The effects are larger for longer breastfeeding durations, and generally persist when the sample is limited to twins to minimize the influence of unmeasured family heterogeneity that changes over time. Taken together, the results of this study suggest that it is too soon to conclude that there are no long-run schooling benefits to breastfeeding.