The assignment of the Econometric Game 2011 was to identify and estimate the causal effect of maternal alcohol consumption during pregnancy on cognitive outcomes of the child, using a genetic marker as instrument.
Mendelian Randomisation is the term used to describe the random allocation of genes from parents to offspring. Recent advances in genotyping have made it possible to obtain genetic information more easily and economists are increasingly interested in the use of genetic variants as instrumental variables to identify the effect of a modifiable non-genetic risk factor on the outcomes of interest. Of course, there may be various reasons why genetic marker may not be a valid instrument. Also, a genetic marker may only be weakly associated with the risk factor.
In the case of the Econometric Game 2011 we contributed to the long and ongoing debate about whether alcohol consumption by the mother during pregnancy has a beneficial or detrimental effect on the development of the foetus. The data set that was analysed contained information about the mother’s alcohol intake during pregnancy and the children’s school test scores at age of 14. A simple regression analysis may not identify the causal effect of alcohol intake due to unobserved factors that affect both maternal alcohol intake and the human capital formation of the child. We had information about the mother on the presence of a genetic marker which has been shown tob be associated with alcohol consumption and may therefore be a good candidate for use as an instrumental variable. This genetic marker may thus enable us to identify/estimate a causal effect of maternal alcohol consumption on the child’s cognitive ability by means of instrumental variables estimation techniques.
The top 3 in 2011:
1. Maastricht University
2. University of Oxford
3. University of Cambridge