Prof. Ana Lucia Kassouf (p)

This study provides the first comprehensive evidence on how hydrological disasters in recent years have affected elementary education outcomes in Brazil, combining innovative data construction with modern causal inference techniques. Using nationally representative administrative records, the analysis links municipal disaster notifications from the Integrated Disaster Information System (S2ID) to both aggregate and individual-level performance data from the Basic Education Assessment System (SAEB) for the period 2013–2017. Two complementary empirical strategies are employed: a municipality-level panel estimated with the staggered difference-in-differences methodology of Callaway and Sant’Anna (2021), and a novel individual-level pseudo-panel built through probabilistic matching and corrected for attrition using inverse probability weighting.
The results show that hydrological disasters significantly reduce student proficiency in Portuguese and Mathematics for 5th-grade students, with effects that persist over the medium run. The impacts are particularly pronounced in the South and Southeast regions, where the frequency and intensity of hydrological events are highest. Crucially, the individual-level analysis reveals learning losses that are substantially larger than those suggested by municipality-level averages, indicating that aggregate models underestimate the true magnitude of disaster-induced educational harm.
The study also uncovers important heterogeneous effects. Boys experience greater losses in Mathematics, girls in Portuguese, and white students show the largest absolute declines across both subjects. Despite these substantial drops, the relative performance gaps between social groups remain largely unchanged. This pattern is consistent with evidence that students with higher baseline achievement levels may experience larger absolute learning losses when exposed to educational shocks, while structurally disadvantaged groups—already constrained by chronic inequalities—exhibit more limited margins for further deterioration.
By combining high-frequency disaster data with standardized learning measurements and rigorous quasi-experimental methods, this research advances the literature on climate shocks and human capital in several ways. It is the first study in Brazil to estimate heterogeneous disaster effects using individual-level educational data and demonstrates the value of integrating staggered DiD estimators with pseudo-panel methods to uncover impacts that would otherwise remain concealed. The findings highlight hydrological disasters as significant drivers of learning inequalities and reinforce the urgency of targeted policies—such as remedial instruction, psychosocial support, and resilient school infrastructure—to protect educational trajectories in a context of increasing climate-related risk.
By documenting how exposure to hydrological events impairs learning and disproportionately affects specific demographic groups, the research provides actionable evidence to inform more equitable and climate-resilient education policies.

Prof. Jiehua Lu, Dr. Yunchen Ruan (p)

Objective: As climate-related disasters become more frequent and severe, understanding their long-term implications for adaptive capacity is increasingly important. This study examines whether adolescent exposure to flood events is associated with life-course cognitive adaptation in later adulthood and considers the broader implications for climate resilience in China, where flood risks are spatially uneven and regionally concentrated.
Methods: We used five waves (2011–2020) of the China Health and Retirement Longitudinal Study (N = 10,321; age 45+). Retrospective reports of adolescent flood exposure were linked to subsequent cognitive performance. Group-based trajectory modeling identified heterogeneous cognitive trajectories over time. Multinomial logistic regression estimated associations between flood exposure and trajectory membership, adjusting for sociodemographic and baseline health characteristics. Inverse probability weighting addressed potential bias from differential survival and attrition. Mediation analyses assessed adult social participation, health behaviors, and intergenerational ties.
Results: Three cognitive trajectories were identified: low-stable (22.6%), moderate-decline (43.0%), and high-stable (34.4%). Light and moderate adolescent flood exposure were associated with a higher likelihood of membership in the moderate-decline and high-stable trajectories relative to the low-stable group. Inverse probability–weighted estimates were consistent. Adult social participation accounted for a modest share of the association (approximately 5–8%), whereas health behaviors and intergenerational ties showed no significant mediation.
Conclusions: Findings suggest that manageable early-life environmental adversity may be associated with stronger long-term cognitive adaptation. Given the geographically concentrated nature of flood exposure in China, these results underscore how spatially differentiated climate risks may contribute to uneven distributions of adaptive human capacities across regions. Understanding such life-course consequences is essential for informing place-based climate resilience strategies.
Keywords: flood exposure; cognitive adaptation; life-course trajectories; climate resilience; regional heterogeneity; China

Dr. Matthias Brachert (p)

This paper studies the short-run and early-career effects of sustained air-pollution exposure during a critical phase of human-capital formation. I exploit a natural experiment generated by the 1982 Soviet oil supply cut to the German Democratic Republic (GDR), which forced the East German regime to substitute imported oil with heavily polluting domestic lignite. The resulting increase in sulfur dioxide (SO₂) pollution was spatially concentrated around lignite-mining and lignite-based energy regions. At the same time, the GDR’s authoritarian institutions strongly limited residential mobility, school choice, and private avoidance behavior, substantially reducing concerns about endogenous sorting.

The empirical analysis proceeds in three steps. First, I document the environmental first stage using newly digitized historical air-quality monitoring data and a county-level panel. Counties located within 60 km of active lignite mining sites experienced a sharp and persistent increase in ambient SO₂ concentrations after 1982. Difference-in-differences estimates imply an increase of about 28.8 µg/m³ (roughly 20 percent relative to pre-shock levels), with no evidence of differential pre-trends. Placebo tests and alternative treatment thresholds confirm the robustness of the pollution effect.

Second, I examine short-run educational outcomes using individual-level university application records from the GDR’s centralized admissions system, which include transcript grades and study preferences. Linking applicants’ school locations to regional pollution exposure, I estimate a difference-in-differences model with rich fixed effects and student controls. Students in more exposed regions experienced a statistically significant deterioration in transcript grades after 1982. The preferred estimate corresponds to about 0.03 grade points (approximately 0.05 standard deviations), with stronger effects in cognitively demanding subjects—especially mathematics and the natural sciences—and larger impacts for male students.

Third, I study early labor-market and tertiary-track outcomes using administrative employment records from the GDR’s GAV register. Cohort-based difference-in-differences estimates show that exposed Abitur cohorts were less likely to enter or complete tertiary education and less likely to enter occupations associated with academic qualifications. The estimated reductions are economically meaningful (about 2.6–2.7 percentage points for tertiary-track outcomes and about 1 percentage point for academic occupations).

Overall, the findings show that a policy-induced pollution shock during late adolescence can generate persistent losses in educational performance and early occupational sorting, highlighting the long-run human-capital costs of environmental degradation in settings with limited scope for behavioral adjustment.

Mr Muhammad Usman (p)

Child malnutrition remains a persistent public health challenge in low-and middle-income countries, driven by complex and spatially uneven environmental and socio-political stressors. This study examines the spatially heterogeneous effects of climate variability and conflict exposure on childhood malnutrition in Pakistan focusing primarily on stunting using high-resolution gridded data from 2001 to 2019. By integrating satellite-derived climate indicators, geolocated conflict events, and socioeconomic covariates, we employ a suite of spatially explicit modelling approaches including Geographically Weighted Regression (GWR), Multiscale GWR (MGWR), and Mixed GWR. Results indicate that the impacts of climatic shocks especially drought intensity and extreme precipitation are highly location-dependent and operate at different spatial scales. Conflict exposure further intensifies these adverse effects in vulnerable and socioeconomically deprived regions, with MGWR revealing stronger fine-scale variations compared to traditional GWR. Mixed GWR identifies a combination of global drivers (e.g., long-term socioeconomic deprivation) and locally varying predictors (e.g., seasonal drought anomalies and conflict proximity). Collectively, the findings highlight substantial spatial heterogeneity in the determinants of child malnutrition and demonstrate that ignoring local variations can mask critical pockets of vulnerability. This study provides a nuanced understanding of how climate and conflict interact across space to influence child health outcomes. The results underscore the need for spatially tailored and multisectoral policy interventions that address both environmental and political fragility to effectively reduce child malnutrition.