http://www.iser.essex.ac.uk/publications/working-papers/iser/2009-16

Recent medical advances have meant that an increasing number of children born at low birth weight (<2,500 grams) and extremely low birth weight (<1,500 grams) are surviving. At the same time, a growing body of epidemiological research has highlighted a strong association between low birth weight and infant mortality, various morbidities (asthma, high blood pressure, lung disease, etc.), cognitive and emotional impairment in childhood, and depression and anxiety in adulthood. Not surprisingly, this body of evidence has had a very strong impact on health policies in many developed and developing countries. Economic research has largely supported the importance of birth weight as a policy target. Using datasets much larger than those available to epidemiologists, and different empirical strategies, most economic studies have found that low birth weight children tend to have lower educational attainment, poorer self-reported health status, higher disability rates, and reduced employment and earnings as compared to their heavier counterparts. A closer look at these findings however reveals several inconsistencies. In some cases the estimated effects are rather large, in others very small. Several papers find a significant impact of birth weight on adult outcomes, but apparently no big effects during the perinatal period or in early childhood. The aim of our study is to provide new evidence on the effects of birth weight on a variety of child outcomes. In order to do so, we use data from the UK Millennium Cohort Study. Our focus is on the early stages of child cognitive and behavioural development, as we believe that if birth weight has large long-term effects on education, income, and earnings, we should observe at least some impact on earlier outcomes. Our empirical strategy differs from what has been done in the previous literature in several aspects. Whereas most existing studies exploit differences in the birth weight of twins or siblings in order to net out the influence of maternal characteristics which are difficult to observe in the data, we use multiple measures of cognitive and non-cognitive achievement (which are available for the same child at the same point in time) to deal with this problem. Secondly, we adopt a production function approach, whereby only individual behaviours and characteristics but not resources or constraints are thought to have a direct effect on the outcomes of interest. This also justifies some of the assumptions required by our empirical setting. Finally, we explore the role of dynamics, i.e. we analyse whether the effect of birth weight on later outcomes acts independently of previous achievements. Our results show that birth weight has a significant effect on male cognitive development at age 3 and on female cognitive and behavioural outcomes at age 3. The magnitude of these effects is however very small. We estimate that a 200-250g increase in birth weight - the most realistic policy target - would increase age 3 cognitive scores by at most 0.03-0.04 standard deviations for boys and at most 0.04-0.05 standard deviations for girls. We also find that birth weight has no significant impact on outcomes measured at age 5 other than through previous outcomes, and in the latter case the dynamics are such that the overall impact of birth weight fades out over time. These results call into question the effectiveness of birth weight as a policy target and highlight the fact that much remains to be understood about the factors which impact on early child development.

In this paper we explore the impact of birth weight on children's cognitive and behavioural outcomes using data from the UK Millennium Cohort Study. In order to deal with the endogeneity of birth weight we propose an eliminant estimator, which exploits the availability of multiple outcomes for the same individual at the same point in time. The results show that birth weight has significant but very small effects on male cognitive development at age 3 and on female cognitive and behavioural outcomes at age 3. We also find that birth weight affects age 5 outcomes only through previous achievements, and that the overall impact fades out over time. These findings call into question the effectiveness of birth weight as a policy target. Working Paper 2009-16 2009-05-14 Emilia Del Bono Institute for Social and Economic Research edelbono@essex.ac.uk John Ermisch Institute for Social and Economic Research ermij@essex.ac.uk