Environmental Justice for Seniors? Evidence from the Superfund Program

Droughts are becoming increasingly common in India. However, most agricultural land remains rainfall dependent. Agriculture employs 50\% of India's workforce who remain vulnerable to rainfall fluctuations. This paper studies the adaptation to droughts of rural households dependent on agriculture. Using a difference-in-difference estimation strategy across drought and non-drought affected regions, I test the effect of drought on annual agricultural labor hours of rural households. Findings suggest that rural households experiencing drought reduce 4.1\% of their fraction of agricultural hours. There is a compositional income shift: the fraction of agricultural income falls and the fraction of non-agricultural income rises with no significant impact on total household income. Results suggest that drought causes households to adapt by diversifying away from agriculture, thereby accelerating the structural change in India. Interestingly, however attachment to agriculture in the form of land ownership mitigates the role of drought on diversification away from agriculture. Cultural obligations and land market transaction costs influence occupational choices in rural India.

Does Drought Accelerate Structural Change in India?

Location-based environmental policies such as the Superfund Program aim to address the disproportionate exposure to environmental hazards of demographic groups. Understanding how the market adjustments post-listing and post cleanup of Superfund sites cause re-sorting of different racial and income groups is crucial to evaluate the policy. A confidential micro dataset of 7 million seniors over 15 years' period (1999-2013) consisting of information about their demographics, presence of over 50 chronic conditions, and individuals' annual addresses up to zip+4 codes. Mapping seniors to Superfund sites and establishing treatment and control areas, following the strategy of Muehlenbachs, Spiller and Timmins (AER, 2015), it is found that post-listing of a Superfund site increases the probability to move out of white individuals by 0.7 percentage points compared to non-whites, for a 4 percent baseline migration rate. Interestingly, post-cleanup, the same pattern is observed but the magnitude falls 0.4 percentage points, throwing some light on the fact that such neighborhoods have some stigma attached and hence delays gentrification. Following the emigrants from Superfund neighborhoods, whites on average are found to locate in places with lower pm2.5 and pm10 concentration compared to non-whites. Each racial type also shows a positive assortative matching to a neighborhood with more individuals of the same race. Understanding how the preference for environmental quality and demographic composition of a neighborhood affects sorting given one's income and race provides important insights about how these policies might undermine the Environmental Justice objectives.


Department of Economics, Arizona State University
P.O. Box 879801 Tempe AZ 85287