Estimating the Effects of Lead Water Pipes on Infant and Fetal Mortality

Estimating the Effects of Lead Water Pipes on Infant and Fetal Mortality 

To identify the effects of lead water mains on fetal and infant death rates, variants on the following generic equation are estimated using cross-sectional data from the year 1900  

If one accepts the plausible assumption that h[ is negatively correlated with both ji and k, it is clear that when h[ is omitted from the regression, as in equation (A.l), the estimated coefficient on lead pipes will reflect the effects of both h and k, and will be biased upward so that ( ) ( ).  

The second strategy explores the correlation between the use of lead service pipes and infectious disease rates. If it were a lack of health consciousness, and not lead water pipes per se, that drove up infant death rates in towns with lead water lines, one would expect to observe a positive correlation between the use of lead water lines and infectious diseases such as scarlet fever, measles, and the like. A positive relationship between lead use and infectious disease might also reflect the fact that exposure to unhealthy lead levels at a young age left children weak and compromised in terms of their ability to fight off infectious diseases. Hence, a positive relationship between lead use and infectious disease rates does not, by itself, prove that it was a lack of health consciousness that caused the elevated rates of infant death in towns with lead lines. However, a negative coefficient, or a very small positive one, on the lead dummy would certainly undermine the case for health consciousness. 

The fourth strategy is to employ instrumental variables. The goal here is to identify an instrument that is correlated with lead use, but not health outcomes. Reasonable instruments are available and these are related to the financing of service pipes, the complexity of urban infrastructure, the number of large industrial users of water, and the ownership 

---