This is the 12th in this year’s series of posts by PhD students on the job market.
The importance of sanitation investment for improving human health in developing countries is widely recognized by policymakers and researchers. Worldwide, 700 million people practiced open defecation in 2016, which is estimated to have caused 432,000 deaths due to diarrhea in that year. To address these issues, developing countries such as India and China have provided significant subsidies for the construction of latrines (toilets). The direct health benefits (e.g., reduced child mortality, improved child growth) of latrine construction have been well-documented.
But little is known about the unintended negative externalities of latrine construction due to poor treatment of fecal sludge, which can offset the direct health benefits. The constructed latrines accumulate a large volume of fecal sludge, which must be emptied periodically by vacuum trucks or by hand. The emptied fecal sludge should then be treated by wastewater treatment plants to disinfect the remaining active pathogens. However, due to insufficient infrastructure, the emptied fecal sludge is instead, in many cases, dumped into rivers, thus polluting the rivers. These water pollution externalities may offset the direct positive effects of reduced open defecation.
My job market paper examines such negative externalities of latrine construction on water quality and health in the context of India's nationwide sanitation policy, the Swachh Bharat Mission (SBM). I find that latrine construction under the SBM increases river pollution by 72%. While the latrine construction reduces diarrheal mortality overall, this positive health effect is two-thirds smaller in areas with lower treatment capacities where water pollution externalities are consequently larger. These findings suggest that unintended water pollution externalities offset positive health effects.
Large-Scale Latrine Construction Policy in Rural India
I look at the case of the SBM, the world's largest program of latrine construction. The SBM has subsidized the construction of over 100 million latrines in rural India since 2014. The amount of subsidy under the SBM is up to 150 US dollars, covering most of the construction cost of latrines. With such a big push, all districts have achieved almost universal latrine coverage by the target date of 2019.
To investigate how this massive latrine construction affects water quality, I use administrative panel datasets on the district-level number of latrines from 2012 to 2019 under the SBM and the water quality of 1,189 monitoring stations along rivers in 337 districts from 2007 to 2019. I combine these with district-level diarrheal post-neonatal mortality estimates to examine the effects on health. The district-level analysis with extensive spatial coverage of these datasets (as shown in figure 1) allows me to examine the negative externalities that extend beyond villages, which was not fully captured in past studies relying on village-level field experiments.
Figure 1: Distribution of Water Quality Monitoring Stations in India
Notes: This figure shows water quality monitoring stations in orange dots, district boundaries in black lines, and rivers in blue lines.
Exploiting Geographical Variation in Soil Characteristics that Affect Latrine Construction
To estimate the causal effects of latrine construction, I mainly adopt an instrument variable (IV) design, exploiting geographical variation in Available Water Capacity (AWC), a proxy for the soil infiltration rate, as an instrument for the number of latrines. Higher soil infiltration rates (lower AWC) increase the risk of groundwater contamination in nearby wells from the fecal sludge accumulated in latrines. To address this risk, an official technical guideline, which became effective since the SBM's inception in 2014, requires either greater distances between latrines and wells or the addition of impervious materials inside latrines in areas with high infiltration rates. So, lower AWC increases the difficulty and cost of latrine construction after the SBM started in 2014. Indeed, I find lower AWC is associated with a smaller increase in latrines during the post-SBM period in the first stage.
To test the assumption of the exclusion restriction, I run reduced-form regressions of the outcomes on the interaction of AWC and year dummies. The exclusion restriction suggests that AWC should not affect the outcomes prior to the SBM policy when the technical guideline was not in effect. Reassuringly, I find no differential effects of AWC on water quality and health until 2013. Conversely, since the policy's inception in 2014, larger AWC has led to an increase in river pollution and a decrease in diarrheal mortality (as shown in figure 2).
Figure 2: Effects of AWC on water quality and health, before and after 2013