Dean E. Spears
The Total Sanitation Campaign built approximately one latrine per ten rural residents of India from 2001 to 2011. This paper uses an annual panel of disaggregated construction data, matched to several sources of household, village, and district data.
- Districts in which more latrines were constructed over this period saw a greater decline in rural infant mortality rates, controlling for other changes, a difference-indifferences result that is strengthened using heterogeneity in an incentive for local government compliance as an instrument for program intensity. This effect is not found for urban IMRs (urban infants were not exposed to the program), nor is a similar correlation with IMR found of district-level intensity of two other government programs from the 2000s.
- Individual data from another source finds a similar decline in infant mortality - quantitatively between the district-level differences and IV results - associated with the TSC latrines built by the first year of a child's life, found using district and year fixed effects and other controls. This effect is through post-neonatal, rather than neonatal,
mortality, and is greater for children who have non-breastmilk food earlier.
- Using a third data source, rural children born in years and districts with more TSC latrines available in the first year of their lives are taller than children born in other years or districts, found using district and year fixed effects; as a placebo test, this is not true of urban children.
Date: March 23, 2012
Time: 11:30 A.M.
Seminar Room 2
Indian Statistical Institute Delhi Centre,
7, S. J. S. Sansanwal Marg,
New Delhi-110016 (INDIA)
View Larger Map