"Reallocation According to Comparative Advantage Under Climate Change: The Role of Land Markets and Coordination in Coastal Bangladesh"
Abstract: Climate change shifts comparative advantage in developing-country agriculture, but whether land markets can deliver the reallocation the new comparative-advantage structure requires is an open question. We study coastal Bangladesh, where sea-level-rise-driven soil salinity has rendered traditional rice agriculture unviable on large areas and made brackish-water aquaculture more profitable, but aquaculture requires consolidating plots beyond the typical scale of land ownership and imposes brackish-water externalities on neighbors. Using a hydro-geological boundary as an instrument for high salinity, we find two things. First, the rental market does deliver large reallocation toward aquaculture: high salinity raises aquaculture land per household by 29 decimals and operational farm size by 28 decimals, achieved through fewer but much larger rental transactions; ownership barely moves. Contrary to the prevailing view that factor-market frictions block reallocation in developing-country agriculture, a suciently large comparative advantage shock overcomes those frictions. Second, the realized reallocation overshoots the individually-optimal benchmark. Among dual-sector adopters, the within-farmer marginal-product wedge between crops and aquaculture is positive (93% of adopters, mean +1.47 log points; robust in sign to instrumental-variables recalibration of the underlying production function elasticities): by these farmers' own production technology, the realized allocation places too much land in aquaculture and too little in crops. The signature of overallocation is concentrated where village-level fundamentals (e.g., soil salinity) do not justify the realized aquaculture share (the wedge declines with own-salinity, rises with village-share), identifying village-level coordination as the operative friction; the brackish-water externality is the leading candidate micro-foundation, though several other village-level frictions produce the same observed patterns. The implied aggregate welfare loss is approximately $1.1 billion per year (range $0.6-1.5B across sensitivity cases). Across coastal developing regions facing sea-level-rise-driven salinity intrusion, the constraint on climate-induced reallocation lies not in factor markets but in the village-level coordination frictions surrounding the adaptive technology.
"Adapting to Moderate, Failing at the Extreme: Newborn Heat Mortality in Bangladesh over Four Decades" (with Anoushka Chopra)
Abstract: Climate-economics models routinely assume that rising income and cooling technology steadily blunt the mortality cost of extreme heat. We show that in one of the world’s most heat-exposed populations adaptation can hold and fail at the same time. Pooling 312,938 births from eight rounds of the Bangladesh DHS, 1994–2022, merged with ERA5 daily-maximum temperature, we let the neonatal heat-mortality slope vary by birth decade. At moderate heat between 30 and 40°C the slope has fallen across cohorts to a value statistically indistinguishable from zero, and a Sun and Abraham (2021) interaction-weighted estimator leveraging the staggered rollout of Bangladesh’s Rural Electrification Board attributes roughly 78 % of this adaptation to the program arrival. At the physiological extreme above 40°C the picture inverts: per ten additional days exceeding 40°C in the 28-day neonatal window, excess mortality rose from 3 per 1,000 for the 1990 cohort to 50 per 1,000 for the 2010 cohort. The same rollout returns a tail estimate indistinguishable from zero: fans cool at moderate heat but not above 40°C, and fans reach 96 % of Bangladeshi households while AC reaches only 1.4 % (BDHS, 2022). Both findings survive a common robustness battery. Cumulative 1990–2010 moderate-heat adaptation averted roughly 420,000 neonatal deaths, worth about $84 billion at the $200K Bangladesh VSL; roughly 328,000 deaths and $66 billion are attributable to the rural electrification program. The tail failure adds about 5,600 excess deaths in the 2010s, and mid-century projections imply between 1,000 and 3,100 excess neonatal deaths per year by 2050, set more by AC coverage than by the emissions pathway.
"From Two Daughters to a Hurried Third: Son Preference, Birth Spacing, and Mortality" (with Syed Manzoor Ahmed Hanifi)
In rural Bangladesh, women whose first two children are both daughters are substantially more likely to attempt a third birth, and to do so sooner, than women with a son. The World Health Organization recommends at least 24 months between pregnancies to protect the next child. Using the plausibly random sex composition of the first two children as an instrument, we estimate the effect of short interpregnancy intervals on third-child mortality in two independent data sources: the Matlab and Chakaria HDSS (34,579 mothers, 2005–2024) and pooled rural Bangladesh DHS rounds (17,946 mothers, 2004–2022). Two-stage least squares with girl-girl as the instrument imply that mothers shifted into a sub-24-month interpregnancy interval face a 9.5 pp higher risk of third-child neonatal death in HDSS (95% CI 2.3–16.6) and 15.8 pp in BDHS (95% CI 1.8–29.9). High-frequency surveillance and nationally representative cross-sectional surveys yield convergent evidence. Combining the two data sources and monetizing at the central Robinson et al. (2019) value-of-statistical-life transfer, the implied annual cost lies in the range $171–$400 million, with an aggregate of $3.4–$8.0 billion in mortality cost over the 2005–2024 sample period.
"Effects of Increase in Decadal Monsoon Onset Variability on Farming: Evidence from Bangladesh"
(with Jared Stolove)
"Migration to the Middle East and Women’s Empowerment in Bangladesh: Evidence on Household Decision-Making and Agency"
(with Cecile Pierre and Frederica Mendonça)
"Adapting to Drinking Water Scarcity in Coastal Bangladesh through Community Level Water Desalination Business"
(with Ahmed Mushfiq Mobarak and Jared Stolove)
"Approaches to Studying Adaptation in Early Climate Change Battlegrounds"
(with Ahmed Mushfiq Mobarak and Matthew Kahn)
“Underinvestment in a Profitable Adaptation Technology: The Role of Market Access”
(with Ahmed Mushfiq Mobarak)