Public Health Efforts and the Decline in Urban Mortality : Reply to Cutler and Miller

This is a rejoinder to a comment by Profs. Cutler and Miller (CM) on a recent paper of ours, “Public Health Efforts and the Decline in Urban Mortality” (NBER WP No. 25027), in which we examine the effect of water filtration on infant and total mortality using the same data as were used by CM (2005). Two main results emerge from our reanalysis of the CM data: (1) correcting their infant mortality counts reduces the estimated effect of filtration by two-thirds, from -43 log points to -13 log points; and (2) using a consistent method of calculating the total mortality rate shrinks the estimated effect of filtration by half, from -16 log points to -8 log points. In this rejoinder, we argue that the much-reduced estimated effect of water filtration on infant mortality is a dramatic and surprising departure from the consensus view in the literature. In addition, we show that the estimated effect of water filtration on total mortality is extremely fragile