Published on Let's Talk Development

When does pollution policy work? The water quality and infant mortality impacts of Mehta vs. Union of India

This page in:
India’s rivers are heavily polluted. According to official estimates, 302 of 445 river stretches fail to meet even bathing criteria ( Central Pollution Control Board [CPCB], 2014). This is known to have a heavy disease burden: each year, 37.7 million Indians are affected by waterborne diseases, 73 million working days are lost, and 1.5 million children are estimated to die of diarrhea alone ( Water Aid, 2008). 
" rel="nofollow">Yavuz Sariyildiz" rel="nofollow">

The holiest of India’s rivers, the Ganga, is one of the most polluted of all. According to recent estimates, each day 501 million liters of industrial effluent are emptied into the Ganga’s waters from 764 “Grossly Polluting Industries” (CPCB, 2013). National legislation and policy efforts to address the challenge of water pollution have widely been characterized as failures (Greenstone and Hanna, 2014).
But on one stretch of the Ganga, meaningful regulation emerged from environmental activism in the Indian judiciary  
In 1987, the lawyer M.C. Mehta brought India’s first-ever environmental public-interest litigation suit to the Supreme Court. The suit targeted the tanning industry, notorious for its water pollution and especially concentrated along the Ganga River in the district of Kanpur. In a landmark ruling, the Court ordered the Kanpur tanneries to clean up or shut down their toxic waste flows into the Ganga River within six months. Our paper assesses the impact of the Mehta vs. Union of India cases on pollution and infant mortality, using monthly data available between 1986 and 2004 (Mehta vs. Union of India was split into two parts: a first that identified tanneries as the defendant, and a second that identified the city government). We compare outcomes in Kanpur to those elsewhere in the Ganga Basin, before and after the verdict, to estimate the impacts of the case on water pollution and infant mortality.
Of the 87 tanneries named in Mehta’s suit, 20 were shut down. Over the subsequent six years, incentives from the Government of India induced many more tanneries in the Jajmau area to establish primary treatment plants and connect to a common effluent treatment plant. Our analysis suggests that these changes spurred real reductions in both water pollution and infant mortality. In particular, the verdict is associated with a 50% drop in both Kanpur’s neonatal (one-month) risk of death and its risk of violating India’s bathing-class standard for water quality.
How much of the Supreme Court ruling’s impact on neonatal mortality can be explained by water quality improvements?
Did infant mortality drop in Kanpur because water quality improved? Or did it do so because, for example, households became aware of river pollution issues and adjusted their water usage accordingly? To shed light on this question, we estimate the direct relationship between water pollution and infant mortality. This task is made challenging by the real-world correlation of pollution with many other things that affect health. Consider that urban districts might both be more polluted and have better health services (for example). Alternatively, poor people may be more likely to be living along heavily polluted stretches of rivers. In these cases, a naïve comparison across locations would conflate the mortality effect of water pollution with the effects of some other, non-pollution factor.
To circumvent this methodological hurdle, we observe that pollution flows downriver, so that river pollution in, e.g., Allahabad affects mortality in, e.g., Varanasi—a city lying 120 km downstream along the Ganga—only through the pollution that is being carried downstream by the river. Consequently, we can then use pollution upstream from a given locality to specifically isolate the impact of local pollution on infant mortality. When we do this, we find that neonatal mortality risk rises 14.4 percentage points when the bathing-class water quality standard is violated.
Given this estimate, we can make a prediction of the Supreme Court ruling’s mortality impacts specifically through the channel of pollution. Comparing it to our initial estimate of the ruling’s impact overall, we are unable to reject that the two are statistically the same. In other words, water pollution reduction has indeed been the major channel through which the ruling has avoided neonatal deaths.
What can we learn from natural experiments in environmental regulation?
The empirical evidence gathered in this analysis shows that the 1987 Supreme Court decision in Mehta vs. Union of India, which primarily targeted the tanning industry in Kanpur district, reduced both surface water pollution and neonatal mortality. In particular, the evidence suggests that water quality improvements explain the majority of mortality reductions. These results corroborate those of other studies that highlight the efficacy of judicial-branch environmental policy in India. Understanding why the Supreme Court has been so effective is key to solving India’s environmental puzzle: persistent pollution problems, even after repeated documentation of the significant social costs of that pollution.


Quy-Toan Do

Co-Director, World Development Report 2023

Samuel Stolper

Postdoctoral Aassociate

Join the Conversation

The content of this field is kept private and will not be shown publicly
Remaining characters: 1000