New Evidence on the Impacts of Birth Order.

• Author: Black, Sandra E.

What determines a child's success? We know that family matters--children from higher socioeconomic status families do better in school, get more education, and earn more.

However, even beyond that, there is substantial variation in success across children within families. This has led researchers to study factors that relate to within-family differences in children's outcomes. One that has attracted much interest is the role played by birth order, which varies systematically within families and is exogenously determined.

While economists have been interested in understanding human capital development for many decades, compelling economic research on birth order is more recent and has largely resulted from improved availability of data. Early work on birth order was hindered by the stringent data requirements necessary to convincingly identify the effects of birth order. Most importantly, one needs information on both family size and birth order. As there is only a third-born child in a family with at least three children, comparing third-borns to firstborns across families of different sizes will conflate the birth order effect with a family size effect, so one needs to be able to control for family size. Additionally, it is beneficial to have information on multiple children from the same family so that birth order effects can be estimated from within-family differences in child outcomes; otherwise, birth order effects will be conflated with other effects that vary systematically with birth order, such as cohort effects. Large Scandinavian register datasets that became available to researchers beginning in the late 1990s have enabled birth order research, as they contain population data on both family structure and a variety of child outcomes. Here, I describe my research with a number of coauthors, using these data to explore the effects of birth order on outcomes including human capital accumulation, earnings, development of cognitive and non-cognitive skills, and health.

Birth Order and Economic Success

Almost a half-century ago, economists including Gary Becker, H. Gregg Lewis, and Nigel Tomes created models of quality-quantity trade-offs in child-rearing and used these models to explore the role of family in children's success. They sought to explain an observed negative correlation between family income and family size: if child quality is a normal good, as income rises the family demands higher-quality children at the cost of lower family size. (1)

However, this was a difficult model to test, as characteristics other than family income and child quality vary with family size. The introduction of natural experiments, combined with newly available large administrative datasets from Scandinavia, made testing such a model possible.

In my earliest work on the topic, Paul Devereux, Kjell Salvanes, and I took advantage of the Norwegian administrative data-set and set out to better understand this theoretical quantity-quality tradeoff. (2) It became clear that child "quality" was not a constant within a family--children within families were quite different, despite the model assumptions to the contrary. Indeed, we found that birth order could explain a large fraction of the family size differential in children's educational outcomes. Average educational attainment was lower in larger families largely because later-born children had lower average education, rather than because firstborns had lower education in large families than in small families. We found that firstborns had higher educational attainment than second-borns who in turn did better than third-borns, and so on. These results were robust to a variety of specifications; most importantly, we could compare outcomes of children within the same families.

To give a sense of the magnitude of these effects: The difference in educational attainment between the first child and the fifth...