28 April 2015

The Birth Order Relationship Revisited: Child Health and Education

The negative association between birth order and educational outcomes has long been well established in the economic and psychological literature. While it is unclear which mechanism explains the observed birth order effect, the literature mainly discusses behavioral channels induced by parents or triggered by siblings. However, some researches have postulated the hypothesis that birth order differences might also in part be mediated by biology. Particularly, for policy makers, the `nature versus nurture question' is a crucial information if birth order differences are unwanted from a welfare point of view and should be addressed accordingly. Albeit to date, evidence on this is missing.

A few studies have pointed attention to the existence of within family differences early in life. Black et al. and Lehmann et al. note that birth order differences are already present at birth. Both studies find, in line with the medical literature, that higher birth order children show better health at birth, for example higher birth weight, which is commonly used as a proxy for in utero conditions as well as neonatal health. Good health at birth, however, is also an indicator for better later-life outcomes and therefore stands in contrast to the negative birth order effect found at older ages.  Until now, however, we have virtually no information on whether these birth differences are also solely behavioral driven or the result of biology and how the health differences evolve throughout childhood.

The purpose of our project is to assess whether birth order effects are already present at birth and to identify the mechanisms through which they might operate. Using Danish Registry Data, that covers the entire population and contains a rich set of variables, we aim to disentangle possible mechanisms of birth order effects that originate from the in utero period or emerge during childhood. We will thereby focus on birth outcomes including anthropometric measures and Apgar score as well as health conditions in early childhood. Observing the children from birth throughout adolescence, we provide important evidence on whether health differences are transmitted to educational outcomes. To our knowledge, we are the first to scrutinize these important interactions. Notwithstanding, in a recent working paper, Lehmann et al follow a similar approach; they examine the evolution of birth order differences in cognitive and non-cognitive test scores over childhood. While they provide important insights into parental time allocation using survey data, we complement their previous findings with our administrative data that allows for the exploration of an even wider scope of channels. More importantly, given the previous literature, we help bring forward a necessary discussion about the nature of birth order effects. This study also enables a vitally important link for any future research that focuses on connecting early life effects with later life outcomes in adulthood.

Our results are twofold. First, we confirm the negative relationship between birth order and educational outcomes. Second and more essentially, we show that the disadvantage of higher birth orders later in life actually starts as a health advantage around birth that reverses at age 5 and turns into a health disadvantage at the beginning of adolescence. Using the richness of the data, we test several specifications to disentangle the effects. Our results suggest that these effects have a strong behavioral dimension. However, our proposed relationship between birth order and health, and the mechanisms that we discuss may still be the product of an unobserved time-variant heterogeneity across siblings and any results found using this relationship should not be interpreted as a causal effect.

Read the publication Birth Order and Health of Newborns: What Can We Learn from Danish Registry Data?

Contact: Anne Ardila Brenøe - University of Copenhagen, Denmark, Ramona Molitor - University of Passau, Germany