Impacts of Extreme Weather Events on Education Outcomes: A Review of Evidence Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Sergio Venegas Marin, Lara Schwarz, and Shwetlena Sabarwal Extreme weather events are increasingly disrupting schooling. Yet, these are underrepre- sented in the climate change literature. Of 15 review articles on the economic impacts of climate change published since 2010, only three mention the impacts of climate change on education. We review available literature on the effects of weather extremes on education. We outline key pathways through which these events impact education outcomes, as well as the magnitude of those impacts. Evidence implies a significant and adverse relation- ship between heat and learning. Studies suggest surpassing a high temperature threshold makes learning difficult and results in learning losses. Across studies, each additional day subject to extreme heat reduces learning. Tropical cyclones, floods, and wildfires precipi- tate school closures, which halt learning. Evidence suggests that one day of school closures leads to one day of learning lost. Weather extremes also negatively impact education out- comes through health , nutrition, poverty, and fragility, among other distal pathways. We discuss the implications of this evidence for policy, including the need to adapt education systems to climate change. Mitigation and adaptation are both urgently needed as extreme weather events become more frequent and severe in the context of climate change. JEL Codes: I20, I25, Q54 Keywords: education, learning outcomes, climate change, extreme weather events, mitigation, adaptation. Introduction Human activity, primarily through the release of greenhouse gases, is indisputably causing climate change. In 2011–2020, global surface temperature was 1.1◦ C above the levels seen in preindustrial times (Bashmakov et al. 2022). The release of green- The World Bank Research Observer © 2024 International Bank for Reconstruction and Development / The World Bank. Published by Oxford University Press https://doi.org/10.1093/wbro/lkae001 39:177–226 house gasses—driven by unsustainable energy use, land use, as well as consump- tion and production patterns—has persistently risen (IPCC 2023). Unsurprisingly, high-income countries bear most of the responsibility for these emissions. Consumption-based emissions data shows that high-income countries are respon- sible for 92 percent of excess global CO2 emissions (Hickel 2020). More specifically, the top decile of households with the highest per capita emissions contributes 34–45 percent of global consumption-based household greenhouse emissions, compared to 13–15 percent contributed by the bottom 50 percent (IPCC 2023). However, the conse- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 quences of these emissions will be more severe and impactful in developing countries (Das Gupta 2014; Sun et al. 2019). Even within countries, the most vulnerable states and communities will be the hardest hit by climate change, exacerbating existing inequities (Kemp et al. 2022). Climate change is increasing the frequency and intensity of extreme events such as floods, droughts, and wildfires, as well as the probability of co-occurring events (Stott 2016; Ebi et al. 2021). The number of disasters has increased by a factor of five over a 50-year period and weather, climate, and water hazards accounted for 50 percent of all disasters over this period (WMO 2021). Approximately 93–95 percent of the world population will experience more than double the current number of hot and dry events by the end of the twenty-first century (Ridder et al. 2022). Children are particularly at-risk to the effects of climate change. Over 99 percent of children around the world are exposed to at least one major climate and environ- mental hazard, shock or stressor and nearly half of the world’s children live in ex- tremely high-risk countries for climate shocks (UNICEF 2021b). In the absence of cli- mate change mitigation, there will be an additional 131,000 child deaths each year by 2030, driven by a higher burden of infectious diseases, food insecurity, and political in- stability (Williams et al. 2021). It is estimated that the education of 75 million children has been disrupted by crises such as conflict and natural disasters. These are projected to increase in frequency and severity with climate change (Theirworld 2018). While the potential effects of climate change on children’s health are relatively well- documented, there has been less systematic attention on the effects of climate change on children’s education outcomes. Out of 15 review articles on the economic impacts of climate change published since 2010, only three mention the impacts of climate change on education.1 On Web of Science, there are five-times as many publications on the health impacts of climate change as on the educational impacts of climate change. The regions of the world that are the most vulnerable to the impacts of climate change, such as sub-Saharan Africa and Southeast Asia, are those that have the least available evidence to quantify how education outcomes are and will continue to be affected. Out of the articles on education and climate, 33 percent (1,903 out of 5,732) are based in the United States and 78 percent (4,467) are from high-income economies (based on World Bank classifications).2 178 The World Bank Research Observer, vol. 39, no. 2 (2024) This paper reviews and summarizes current evidence on the impacts of extreme weather events on education outcomes. Focusing on extreme heat, tropical cyclones, floods, droughts, and wildfires, research on both the direct and indirect impacts of these extreme weather events on short-term and long-term educational indicators is reviewed. Implications of the evidence for the education sector are discussed. The pa- per also highlights evidence gaps and potential opportunities for future work on this topic. Extreme weather events will only become more severe under climate change; we must understand their impacts on education systems, schooling, and long-term eco- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 nomic context to best prepare and protect the populations and countries that are most vulnerable. The paper is organized as follows. We first present a framework summarizing the di- rect and indirect ways in which extreme weather events impact education outcomes. Next, we summarize evidence on the impact of the four direct channels (extreme heat, tropical cyclones/floods, droughts, and wildfires) on learning, enrollment, dropouts, and long-term outcomes. After this, we present evidence on the indirect channels through which extreme weather events affect education outcomes. Next, we discuss how to protect and leverage the education sector in the face of climate change. The final section concludes. Impacts of Climate Change on Education: Overview of Framework Climate change erodes education outcomes by impacting learning, dropouts, and long- term outcomes (fig. 1). Extreme weather events have proximate, or direct, effects on education outcomes through compromised school access, school closures, destruction of school structures, and impacts on school environment (see table 1 for details on key papers included in the review). There are also more distal, or indirect, effects of these climate extremes on educa- tion outcomes. Extreme weather can lead to economic shocks, produce poor air qual- ity, increase conflict and food insecurity, and drive higher migration and displacement. Increasing extreme weather and climate events have already driven millions of people towards acute food insecurity and reduced water security, with the largest adverse im- pacts observed among the most vulnerable locations and communities (IPCC 2023). Climate change will produce more conflict and violence, especially in low-income re- gions of the world (Akresh 2016). For example, climate-driven food insecurity and wa- ter scarcity are projected to increase with increasing global warming, interacting with non-climatic risk drivers such as more frequent pandemics, conflicts associated with competition for resources (land, water), and forced displacements (IPCC 2023). These will exacerbate disruptions to education services, reduce demand for schooling due to coping mechanism to account for economic losses at the household level, and reduce the readiness of students due to health and nutritional shocks (Caruso et al. 2024). Poverty, food insecurity, and stress make populations more vulnerable to the effects Venegas Marin et al. 179 Figure 1. Impacts of Climate Change on Education Outcomes Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Source: Author’s illustration. of extreme weather and measures to address these underlying vulnerabilities will not only address some of the root causes of educational disparities but can also increase resilience to climate hazards. Extreme Heat and Education Outcomes Extreme heat poses a threat to children and their education outcomes. Around 30 per- cent of the world’s population is exposed to extreme heat conditions and future pop- ulation exposure is projected to increase by 48 percent to 74 percent under differing climate change scenarios (Mora et al. 2017). Over one-third of children globally are currently exposed to heatwaves (UNICEF 2021a). The increasing exposure to extreme heat poses a notable challenge for the education sector. A growing body of evidence suggests that high temperatures can adversely impact both short-term and long-term academic performance. 180 The World Bank Research Observer, vol. 39, no. 2 (2024) Table 1. Key Literature on the Impacts of Extreme Weather on Educational Outcomes Title Author Venegas Marin et al. Exposure Journal, Year Country Student population Study type Key finding High Hot Temperature and High-Stakes United States high school students Observational Taking an exam when the temperature temperature Performance is 90◦ F results in 13 percent of a R. Jisung Park standard deviation lower exam Journal of Human Resources, 2022 performance relative to a more optimal 75◦ F. High Learning is inhibited by heat 58 countries global- 15-year-olds (PISA), Observational For each additional hot day (defined as temperature exposure, both internationally and US- 3rd to 8th grade (SEDA) a day above 80◦ F) in the three years within the United States preceding the exam, learning is lowered R. Jisung Park, A. Patrick Behrer & by 0.0018 standard deviations. Joshua Goodman Nature Human Behavior, 2020 High Heat and Learning United States high school (PSAT) Observational For 3 additional hot days (defined as 3 temperature R. Jisung Park, Joshua Goodman, days above 90◦ F) in the year prior to the Michael Hurwitz, Jonathan Smith exam, learning is lowered by 0.002 American Economic Association, 2020 standard deviations. High Temperature and high-stakes China high school (National Observational A 1◦ C increase in temperature during temperature cognitive performance: Evidence College Entrance the exam period decreases total test from the national college entrance Examination) scores by 2.91 percent of a standard examination in China deviation. Joshua Graff Zivin, Yingquan Song, Qu Tang, Peng Zhang Journal of Environmental Economics and Management, 2020 181 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued 182 Title Author Exposure Journal, Year Country Student population Study type Key finding High The relationship between Global Elementary to high school Review Performance on psychological tests and temperature classroom temperature and level school tasks can be expected to increase children’s performance in school on average by 20 percent if classroom Pawel Wargocki, Jose Ali temperatures are lowered from 30◦ C to Porras-Salazar, Sergio 20◦ C. Temperature for optimal Contreras-Espinoza performance is lower than 22◦ C in Building and Environment, 2019 temperate climates. High Effects of Heat on Mathematics Vietnam high school (National Observational An increase of 1◦ F results in an temperature Test Performance in Vietnam college and university approximate 0.006 standard deviation Tien Manh Vu entrance examinations) loss of in standardized test scores. Asian Economic Journal, 2022 High Heat and learning in elementary United States elementary and middle Observational Increasing average temperature levels temperature and middle school school and particularly hot days reduce Travis Roach, Jacob Whitney student learning and achievement- each Education Economics, 2020 day above 100◦ F decreases average student achievement by 2.3 percent. High Classrooms’ indoor Global higher education (college Review Thermal neutral sensation will occur at temperature environmental conditions and University) different indoor temperatures, between affecting the academic 19.5 and 23.3◦ C. This will vary by achievement of students and outdoor temperature and period of teachers in higher education: A transition to classroom temperature. In systematic literature review one study, students’ cognitive Henk W. Brink, Marcel G. L. C. performance peaked at 23◦ C and Loomans, Mark P. Mobach, Helianthe declined by 48 percent at a higher S. M. Kort temperature of 30◦ C. Indoor Air, 2020 The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding Venegas Marin et al. High Temperature and Human Capital India primary and secondary Observational For each additional hot day (defined as temperature in India school a day above 29◦ C) in the year prior to Teevrat Garg, Maulik Jagnani, and Vis the exam, learning is reduced by 0.003 Taraz standard deviations (in math) and 0.002 Journal of the Association of deviations (in reading) Environmental and Resource Economists, 2020 High The effects of summer heat on Korea high school (Korean college Observational For each hot day (defined as a day with a temperature academic achievement: A cohort entrance exam) maximum temperature above 34 C) in analysis the summer preceding the exam, Hyunkuk Cho learning in math and English was Journal of Environmental Economics reduced by 0.0042 and 0.0064 standard and Management, 2017 deviations. High Weather and high-stakes exam Brazil high school (Brazil college Observational No strong effect of temperature on exam temperature performance: Evidence from entrance exam) performance was observed, with a 1◦ C student-level administrative data increase in dry bulb temperature in Brazil improving z-scores by Xiaoxiao Li, Pankaj C. Patel three-thousandths of a standard Economics Letters, 2021 deviation. High Temperature, effort, and Brazil high school (Brazil college Observational One standard deviation increase in temperature achievement: Evidence from a entrance exam) temperature during the exam decreases large-scale standardized exam in the average exam score by 0.036 of a Brazil standard deviation. Ana Paula Melo, Mizuhiro Suzuki Working paper, 2021 183 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 184 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding High Temperature and Low-Stakes China population older than 10 Observational High temperature on test day (above temperature Cognitive Performance years 32◦ C on the test date relative to a day Xin Zhang, Xi Chen, Xiaobo Zhang within 22◦ C–24◦ C) lowers math scores Journal of the Association of by 0.066 standard deviations. Environmental and Resource Economists, 2024 High Decreased humidity improves China college students Experimental At constant relative humidity, temperature cognitive performance at extreme increasing indoor temperature from high indoor temperature 26◦ C to 39◦ C increased reported health Xiaoyu Tian, Zhaosong Fang, Weiwei symptoms of dry throat, dry skin and Liu headaches, dizziness, difficulty in Indoor Air, 2020 thinking and concentrating clearly, fatigue, and decreased well-being and mood. This also decreased accuracy on cognitive tests. High Investigation of students’ China university students Experimental No significant difference in short-term temperature short-term memory performance memory performance observed for and thermal sensation with heart changes in temperature and correlated rate variability under different color temperature. environments in summer Yalong Yang, Lin Hu, Rui Zhang, Xulai Zhu, Mingyue Wang Building and Environment, 2021 The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued Title Author Venegas Marin et al. Exposure Journal, Year Country Student population Study type Key finding High Effects of different classroom Italy university students Experimental Cognitive C-score scores were lower temperature temperatures on cardiac during day 2 (hotter day) in a two day autonomic control and cognitive experiment. Student’s thermal comfort performances in undergraduate was lower during Day 2 compared to students Day 1. F Barbic, M Minonzio, B Cairo, D Shiffer, A Dipasquale, L Cerina, A Vatteroni, V Urechie, P Verzeletti, F Badilini Physiological Measurement, 2019 High Effects of a cool classroom Italy university students Experimental Greater cognitive performance temperature microclimate on cardiac observed during the COOL trial was autonomic control and cognitive associated with a lower cardiac performances in undergraduate sympathetic and greater cardiac vagal students modulation. Franca Barbic, Maura Minonzio, Beatrice Cairo, Dana Shiffer, Luca Cerina, Paolo Verzeletti, Fabio Badilini Martino Vaglio, Alberto Porta, Marco Santambrogio, Roberto Gatti, Stefano Rigo, Andrea Bisoglio, Raffaello Furlan Science of the Total Environment, 2022 185 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 186 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding High Reducing classroom temperature Costa Rica elementary school students Experimental An intervention that reduced classroom temperature in a tropical climate improved the temperatures showed that the thermal comfort and the 11-year-old children had a significantly performance of elementary school higher speed on the language and pupils logical-thinking tasks at the lower Jose Ali Porras-Salazar, David P. temperature, while the less able pupils Wyon, Beatriz Piderit-Moreno, Sergio performed better on all tasks at the Contreras-Espinoza, Pawel Wargocki lower temperature. There were no Indoor Air, 2018 significant effects on accuracy. High Climate warming causes declines Cameroon School-aged children (5–19) Observational Climate has decreased plantain yields in temperature in crop yields and lowers school Cameroon which is associated with a attendance rates in Central Africa decrease in school attendance. Trevon L. Fuller, Paul R. Sesink Clee, Kevin Y. Njabo, Anthony Tróchez, Katy Morgan, Demetrio Bocuma Meñe, Nicola M. Anthony, Mary Katherine Gonder, Walter R. Allen, Rachid Hanna, Thomas B. Smith Science of the Total Environment, 2018 High Climate variability and Ethiopia Children aged 12–16 Observational Milder temperatures during all seasons temperature educational attainment: Evidence and greater rainfall during the summer from rural Ethiopia agricultural season in early life are Heather Randell, Clark Gray associated with an increased likelihood Global Environmental Change, 2016 of children completing any education. The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding High Effects of Classroom Ventilation United States Elementary school students Observational Each 1◦ C decrease in classroom Venegas Marin et al. temperature Rate and Temperature on temperature increased test scores by Students’ Test Scores 12–13 points. Ulla Haverinen-Shaughnessy & Richard J. Shaughnessy PloS one, 2015 Cyclones, Integration of face-to-face and Brazil Biology under-graduate Observational On flood days, student participation hurricanes, virtual classes improves test was lower, but have virtual classes floods scores in Biology undergraduate attenuated these effects. courses on days with flooding in Brazil Santana OA, Silva TP, Oliveira GS, Silva MM, Inacio ED, Encinas JI. Acta Scientiarum. Education, 2013 Cyclones, The impacts of natural disaster on Thailand Students grade 6–12 Observational Floods had negative effects on test hurricanes, student achievement: Evidence scores for students in grade 6 and 9, but floods from severe floods in Thailand not grade 12m except for social studies. Kawin Thamtanajit Changes in test score from flooding The Journal of Developing Areas, 2020 ranged from 0.03 to 0.11 standard deviations, depending on the subject and level. Cyclones, Medical Education in Post-Katrina United States medical students Descriptive There are long term effect on Tulane hurricanes, New Orleans University School of Medicine and floods A Story of Survival and Renewal Louisiana State University School of N. Kevin Krane, Richard P. DiCarlo, Medicine at New Orleans (LSU) Marc J. Kahn following hurricane Katrina. JAMA, 2007 187 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 188 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding Cyclones, The impact of extreme weather on Philippines undergraduate and graduate Observational Student online participation decreased hurricanes, student online learning students by − 20 percent due to typhoons. floods participation Ezekiel Adriel D. Lagmay & Maria Mercedes T. Rodrigo Research and Practice in Technology Enhanced Learning, 2022 Cyclones, Academic and Student Affairs United States community and technical Descriptive There were long-term effects on the hurricanes, Issues Post Hurricane Katrina colleges educational system post-Katrina and floods Camille Jarrell, Raymonda Dennis, there is a needs for crisis-management Marian Jackson & Cynthia A. Kenney programs in college and university Community College Journal of settings. Research and Practice, 2008 Cyclones, The storm after the storm: the United States school districts Descriptive There are challenges that prolong the hurricanes, long-term lingering impacts of impacts of disasters and impede floods hurricanes on schools recovery in schools. There is a need to Cassandra R. Davis, Sarah R. Cannon, identify the best ways to support Sarah C. Fuller schools, months to years after an event. Disaster Prevention and Management, 2021 The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding Venegas Marin et al. Cyclones, Hidden disasters: Recurrent Philippines school communities Descriptive There are severe impacts of small scale hurricanes, flooding impacts on educational (students, teachers + staff) floods to school communities, floods continuity in the Philippines particularly for the most marginalized Jake Rom D. Cadag, Marla Petal, students. Approximately 7–12 days of Emmanuel Luna a, J.C. Gaillard, school are lost due to small scale floods Lourdes Pambid, Genia V. Santos every year. International Journal of Disaster Risk Reduction, 2017 Cyclones, School hazard vulnerability and Philippines elementary and secondary Observational Provinces located near inland bodies of hurricanes, student learning public school students water are vulnerable to floods and those floods Clarissa C. David, Sheryl Lyn C. in the eastern side of the country are Monterola, Antonino Paguirigan Jr., most vulnerable to typhoons. The Erika Fille T. Legara, Anjali B. Tarun, repeated use of school structures as Rene C. Batac, Jeriesa P. Osorio evacuation centers has negative impact International Journal of Educational on school performance. Research, 2018 Cyclones, School vulnerability to disaster: United States school districts Observational School districts located in coastal hurricanes, examination of school closure, counties on Hurricane Ike’s path have floods demographic, and exposure an increase in the number of school factors in Hurricane Ike’s wind closure days. swath A.-M. Esnard, B. S. Lai, C. Wyczalkowski, N. Malmin & H. J. Shah Natural Hazards, 2017 189 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued 190 Title Author Exposure Journal, Year Country Student population Study type Key finding Cyclones, The impact of flood disasters on Zimbabwe school children Observational Floods cause loss of learning hours, loss hurricanes, child education in Muzarabani of qualified personnel, outbreaks of floods District, Zimbabwe waterborne diseases, high absenteeism Chipo Mudavanhu and low syllabus coverage leading to Jàmbá: Journal children’s poor academic performance. of Disaster Risk Food insecurity, being withdrawn from Studies, 2014 school and sometimes forced into early marriages are also described as challenges faced by children. Cyclones, The Effects Of 2010 Flood On Pakistan school infrastructure Descriptive Infrastructure along with furniture and hurricanes, Educational Institutions And fixture and office record of 23 floods Children Schooling In Khyber schools in two districts were damaged. Pukhtoonkhwa: A Study Of The institutions were closed for three Charsadda And Swat Districts and more than three months either due Saif-ur-rehman Saif Abbasi & Bilal to damage caused by flood or using Shaukat these institutions as shelter homes for International Journal of flood victims. Environment, Ecology, Family and Urban Studies (IJEEFUS), 2013 Droughts The impact of drought on human India Primary and secondary Observational Droughts led to a decline of 4.14 percent capital in rural India school in math scores and 2.67 percent in Kuhu Joshi reading scores of affected children. Environment and Development Economics, 2019 The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding Droughts Impact of a Severe Drought on Zimbabwe Girls in primary school Observational A 4 percent decline in math scores was Venegas Marin et al. Education: More Schooling but observed for affected girls in Zimbabwe Less Learning but literacy scores were not affected. Ardyn Nordstrom, Christopher However, the probability that students Cotton advance in school increased likely due SSRN, 2020 to lower opportunity costs to education Droughts Drought of opportunities: India Primary school Observational Higher wages increase human capital Contemporaneous and long-term investment in early life (in utero to age impacts of rainfall shocks on 2) but decrease human capital from age human capital 5 to 16. When rainfall is higher, children Manisha Shah, Bryce Millett switch out of school into productive Steinberg work. Journal of Political Economy, 2017 Droughts Rainfall shocks, cognitive Kenya Girls in primary school Observational Past rainfall shocks have substantial development and educational negative effects on the cognitive attainment among adolescents in development and educational a drought-prone region in Kenya achievement of girls. The probability of Laura Nübler, Karen Austrian, John correctly answering all four reading A. Maluccio, Jessie Pinchoff questions was reduced by 13.0 Environment and Development percentage points for girls with a Economics, 2021 rainfall shock at age 0, i.e., during the first year of life. The probability of ever having enrolled in school was reduced by 16.9 percentage points and of having enrolled by age 8 by 20.7 percentage points. Years of schooling completed was reduced by 0.8 grades. 191 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 192 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding Droughts How Do Early-Life Shocks Interact Colombia High school exam Observational Children exposed to 8 months of with Subsequent “normal” weather conditions in utero Human Capital Investments? and early childhood experiences a 5.0 Evidence from Administrative percent increase in the probability of Data not dropping out of school and a 0.05 SD Valentina Duque, Maria increase in exam scores. Rosales-Rueda, Fabio Sanchez Working paper, 2019 Droughts Impact of weather shocks on Colombia middle and high school Observational Rainfall shocks reduce math scores by educational outcomes in the 3.95 and reading scores by 4.24. municipalities of Colombia Paola Palacios, Libardo Rojas-Velásquez International Journal of Educational Development, 2023 Wildfire Smoke Lower test scores from wildfire United States grades 3–8 Observational Relative to a school year with no smoke, smoke exposure average cumulative smoke-attributable Jeff Wen, Marshall Burke PM2.5 (surface particulate Nature Sustainability, 2022 matter < 2.5 μg m − 3) exposure during the school year (∼35 μg m − 3) reduces test scores by ∼0.15 percent of a standard deviation. The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Table 1. Continued Venegas Marin et al. Title Author Exposure Journal, Year Country Student population Study type Key finding Wildfire Smoke Do Wildfires Harm Student United States grades 3–8 Observational Wildfire smoke plumes significantly Learning? lowers test scores- a 10-day increase in Ge Wu smoke exposure within 1 year prior to Haslam College of Business Working tests decreases students’ ELA test scores Paper, University of Tennessee, by 0.003 standard deviations and math Knoxville, 2022 scores by 0.004 standard deviations. Wildfire Smoke Impact of short school closures United States schools Observational Wildfires are responsible for the (1–5 days) on overall academic majority of school closures in California. performance of schools in Wildfires generate significant negative California impacts on academic performance Rebecca K. Miller & Iris Hui among younger students. Scientific reports, 2022 all Climate Change and Schools: United States schools Descriptive Flooding, hurricanes, tornadoes, and Implications for Children’s Health wildfires are challenging communities and Safety to re-evaluate whether their schools Stephanie Chalupka and Laura provide a safe, healthy environment. Anderko Creative Nursing, 2019 193 Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 194 Table 1. Continued Title Author Exposure Journal, Year Country Student population Study type Key finding all Assessment of vulnerability and global schools Review Damaged infrastructure causes school resilience of school education to closures for extended periods, which climate-induced hazards: a review can produce discontinuity in education. Anushree Pal,Takuji W. Tsusaka,Thi There is a need for policies on how to Phuoc Lai Nguyen &Mokbul Morshed implement Comprehensive School Ahmad Safety Framework and build schools Development Studies Research, 2023 resilience. all Climate Changes Affect Human Conceptual All ages Review From direct effects on health, nutrition, Capital framework and wellbeing, and indirect effects on Germán Caruso, Inés de Marcos & economic systems, markets, and Ilan Noy through damage to infrastructure, Economics of Disaster and Climate climate change impacts human capital Change across the life cycle. 2024 Source: Authors’ analysis based on review of selected literature. Note: Included impacts are not exhaustive. The World Bank Research Observer, vol. 39, no. 2 (2024) Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Elevated temperature in classroom environments has acute effects on the physiol- ogy of students, increasing heart rates and affecting perception and spatial orientation (Brink et al. 2021). These physiological responses have been linked to reduced accuracy and speed in cognitive tests, translating into diminished learning outcomes (Porras- Salazar et al. 2018; Barbic et al. 2019; Tian et al. 2021; Barbic et al. 2022; Vu 2022). Cumulative exposure to higher temperatures or “hot” days in the months/years prior to an exam can also impact learning. Rather than evaluating the acute effects of heat, these studies consider the medium or long-term negative effects of exposure to Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 extreme heat on school days (Cho 2017; Garg et al. 2020; Park et al. 2020). In the medium or long-term, heat not only affects learning, but can also trigger lower attendance and higher dropouts. Extreme heat will disproportionally affect the poorest regions. Warmer and lower re- source settings are facing higher exposures to extreme heat conditions and as a result experiencing the greatest burden on education outcomes. Each hot day can also have different effects on populations across and within countries, as the socioeconomic con- ditions and resources of schools and students will affect resilience to these stressors. For example, hot days may not have the same effect on students attending a school with air conditioning as on students in a school with no electricity and improper ventilation. Impacts on Learning Exposure to high temperatures has a negative impact on learning outcomes. A num- ber of studies have evaluated how exposure to high temperatures in the year(s) prior to standardized tests affect scores. In Brazil, an increase of 1◦ C during the two years prior to a national assessment translates into ∼0.03SD losses in test scores (Schady et al. 2024). Park et al. (2020) found that learning is reduced by 1 percent for every 1◦ F increase in temperature in the school years preceding exams, using scores from the Preliminary Scholastic Aptitude Test (PSAT), a practice college entrance exam in the United States (Park et al. 2020). Roach and Whitney (2022) also evaluated the ef- fects of average maximum temperature on English/Language Arts and Math test scores from the Stanford Educational Data Archive (SEDA) for students in third grade through eighth grade across the United States (Roach and Whitney 2022). They find that a 1◦ F increase in average maximum temperature leads to up to a 4.71 percent reduction in mean test scores (Roach and Whitney 2022). Larger effects are observed when considering the effects of extreme heat days on learning. Several studies have shown that for every additional day of extreme heat, at least one day of learning is lost (Cho 2017; Garg et al. 2020; Park et al. 2020; Park et al. 2021; Roach and Whitney 2022). Across 58 countries participating in the Programme for International Student Assessment (PISA) between 2000 and 2015, Park et al. (2021) found that exam scores decreased by 0.18 percent of a standard deviation for every additional hot day (above 26.7 ◦ C/80 ◦ F) in the three years prior to the exam, equivalent Venegas Marin et al. 195 to 1.08 days of learning lost.3 This was driven by high temperatures on school days (on school days, 0.22 percent change equivalent to 1.33 days lost) and affected primarily poorer countries. Besides the global analysis, only a few studies have considered the effects of cumulative exposure to high temperatures on academic achievement outside of the United States. Cho (2017) used city-level temperature data to evaluate the effects of hot days during the summer on test scores for the Korean college entrance exam, which occurs in November (Cho 2017). The study found that math and English scores decreased by 0.0042 (equivalent to 2.52 days of learning lost) and 0.0064 (equivalent to Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 3.84 days of learning lost) standard deviations, respectively, with every additional day with a maximum daily temperature over 34◦ C relative to a day in the 28–30◦ C in the summer months. Garg et al. (2020) used survey data on educational achievement in primary school children in India to evaluate the effect of the number of hot days in the calendar year prior to the year of the test on scores, finding that math and reading test performance was reduced by 0.03 and 0.02 standard deviations, respectively, equivalent to 1.2 to 1.8 days of learning lost, for each additional day with average daily temperature more than 29◦ C relative to 15◦ –17◦ C (Garg et al. 2020). Extreme heat is also found to hinder performance on the day of the exam. Across six studies considering temperature on the exam day, the effect of a 1◦ C increase in tem- perature during the exam lowered test scores up to 2.91 percent of a standard deviation (Zivin et al. 2020; Melo and Suzuki 2021; Li and Patel 2021; Park 2022; Vu 2022; Zhang et al. 2024). This has consequences on student achievement and academics, which could be particularly problematic for high-stakes exams where one’s future stands on the line. Zivin et al. (2020) studied how hot weather affects students’ performance in China’s National College Entrance Examination (NCEE). They discovered that for every 1◦ C rise in temperature during the two-day exam period, test scores dropped by 2.91 per- cent of a standard deviation (Zivin et al. 2020). Melo and Suzuki (2021) conducted a similar analysis in Brazil, considering impacts of daily municipality-level temperature estimates on scores on the National High School Evaluation Exam (ENEM), which is used for students’ admission to universities. They find that a one standard deviation increase in temperature (equivalent to 3.679◦ C) during the exam decreases scores by 0.036 of a standard deviation (Melo and Suzuki 2021). Vu (2022) also studied effects of average temperature on test days on scores in the Vietnamese national university and college entrance examinations in 2009 and found that a 1◦ F increase on the test day was associated with an approximate 0.006 SD decrease in standardized test scores (Vu 2022). Students in rural areas and women were found to be particularly vulnerable to these effects (Vu 2022). Park (2022)evaluated effects of exam-day ambient temper- ature on high stakes exams in New York city and found a 13 percent of a standard de- viation lower exam score when the temperature was 90◦ F as compared to 75◦ F (Park 2022). A study in China found that temperatures above 32◦ C on the test date, relative 196 The World Bank Research Observer, vol. 39, no. 2 (2024) to days within 22◦ C–24◦ C, decrease math scores by 0.066 standard deviations, which they translate to be equivalent to losing 0.23 years of education (Zhang et al. 2024). The human body’s reaction to heat stress is at the core of these impacts. A body’s physiological response to heat stress can produce vasodilation, an increased blood flow to the skin, putting increased strain on the heart which affects the body’s ther- moregulatory capacity (Ebi et al. 2021). Exposure to higher temperatures can also pro- duce changes in heart rate and respiratory rates. A study on emergency department visits of children in the United Kingdom found an increase of approximately 10 beats Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 per minute for every degree Celsius increase in body temperature (Davis et al. 2021). The effects of high temperature on respiratory rates are less clear with some stud- ies finding no clinically meaningful effects on respiratory rates (Heal et al. 2022) and others showing an increase in respiratory rate of up to 2 breaths per minute per de- gree Celsius increase in body temperature (Davies and Maconochie 2009). These phys- iological responses to high temperatures, as well as the discomfort and distraction of being in a hot environment, can affect students’ learning and academic achieve- ment. Raising both skin and core temperature can increase strain on the cardiovascu- lar system, producing faster reaction times and decreased accuracy in cognitive tests (Simmons et al. 2008). A systematic review on ambient air temperature and cognitive performance finds that hot temperatures decrease performance, leading to a 5 percent decrease in accuracy and 3 percent decrease in speed in cognitive tasks (Yeganeh et al. 2018). Research studying the indoor built environment and human comfort demonstrates the association between indoor temperature and thermal comfort. This evidence uses physiological indicators such as heart rate, blood pressure, heart activity, brain activ- ity, skin conductivity and oxygen saturation to evaluate these effects. An experimental study considering 60 different environmental conditions with varying combinations of air temperature, relative humidity, and air speed found that air temperature had the strongest effect on heart rate variability (HRV), which is associated with thermal comfort (Liu et al. 2008; Zhu et al. 2018; Yang et al. 2021). A study evaluating cogni- tive performance under hot environmental conditions found that results for cognitive tasks decreased as heart rate variability increases (Zhu et al. 2018). A review of the ef- fects of acoustic, thermal, and illumination conditionson human perception and per- formance, including both experimental studies and field experiments in indoor envi- ronments, identified that temperature level considered satisfactory for human comfort is between 21◦ C and 27◦ C (Yang and Moon 2019; Yang et al. 2021). Experimental studies have explored the effect of classroom temperature on stu- dent cognitive outcomes, measured by memory, speed, and accuracy in cognitive tests. Across five out of seven experimental studies considering the effect of class- room temperature on cognitive performance, for every 1◦ C increase in classroom temperature, student performance decreases by an estimated range of 2 percent to 12 percent (Porras-Salazar et al. 2018; Barbic et al. 2019; Wargocki et al. 2019; Venegas Marin et al. 197 Brink et al. 2021; Barbic et al. 2022).4 A review and meta-analysis compared effects of temperature on schoolwork and office work and found that the optimal tem- perature for schoolwork is lower and that effects are greater in a school setting. They suggest that children may have different thermal perceptions and find that children prefer classroom temperatures 2–3◦ C lower than adults in office spaces (Wargocki et al. 2019). Barbic et al. (2022) found that even under minimal thermal discomfort, university students in a cool (18.4 ± 0.4◦ C) classroom had a lower heart rate variability than students in a neutral (21.5 ± 0.8◦ C) temperature classroom, which Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 also produced improvements in short-term memory, verbal ability, and overall cog- nitive scores (Barbic et al. 2022). A study by some of the same authors also showed that lower classroom temperatures in a two day-study (Day 1: 22.4◦ C ± 0.1◦ C; Day 2: 26.2◦ C ± 0.1◦ C) showed decreased thermal discomfort and a shift in sympathetic pre- dominance and higher short-term memory, verbal ability, and overall cognitive scores (Barbic et al. 2019). In China, higher classroom temperatures increased reported health symptoms of dry throat, dry skin, headaches, dizziness, difficulty in thinking and con- centrating clearly, fatigue, and decreased well-being and mood (Tian et al. 2021). Most concerningly, because rising temperatures is a slow and gradual process, even slight increases in heat have a significant cumulative impact. A forthcoming study from Brazil delves into the connection between heat and learning, providing valuable in- sights (Schady et al. 2024). The study finds that in the hottest 10 percent of munici- palities, maximum daily temperatures experienced by schools rose more quickly (at a rate of about 0.6 degrees Celsius per decade) than in the other 90 percent. In these mu- nicipalities, which are also the most disadvantaged, students lost about 1 percent of learning per year due to rising temperatures. If one assumes that learning is entirely cumulative with each year’s learning building on those before, and that each year 1 percent of learning is lost, by the time a student completes grade 12, that student will have lost about 1.5 years of learning. If by contrast, one assumes a more realistic mix of learning being cumulative in the early years and more independent in the later years, by the time a student completes grade 12, that student will have lost about 0.66 years of learning. In essence, the study finds that a child who enters 1st grade today in a mu- nicipality already experiencing high temperatures will lose between 0.66 and 1.5 years of learning by the time they graduate from 12th grade. Given the assumptions, the esti- mate of 1.5 years of learning lost is the unlikely upper bound of possible impacts, while 0.66 is a more reasonable estimate that signals the magnitud of the problem. The combination of these strands of research offers an understanding of the com- plex relationship between heat and learning. Studies that look at increases in the av- erage temperature in the year(s) prior to an exam find relatively small impacts while studies that look at the incidence of extreme heat on school days find larger impacts. Experimental studies add texture to these findings. They suggest that it is surpassing a temperature threshold that makes learning more difficult and results in learning losses, 198 The World Bank Research Observer, vol. 39, no. 2 (2024) rather than a linear relationship between temperature and learning. This implies that there are temperature levels at which it is just difficult to focus, so it is only when those levels are reached that learning is inhibited. Increasing temperature by 1 degree may not have adverse effects depending on the context, but if that extra degree pushes the temperature to a level that makes learning difficult, then learning losses happen in the absence of any adaptation measure. While the impact of individual hot days may be perceived by some to be small, they compound over time resulting in significant learn- ing losses. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Impacts on Enrollment, Dropouts, and Long-Term Outcomes Decreased levels of learning lead to higher dropout rates. When children are exposed to extreme heat, they learn less. This has implications that go beyond learning. Several studies have shown that lower student performance is linked with a higher probability of dropping out, and vice versa. In Senegal, a one standard deviation improvement in a grade 2 test score is associated with a 22-percentage point increase in the probability of completing grade 6 (Glick and Sahn 2010). Similarly, when a test score is one standard deviation higher, students in Ethiopia, India, Peru, and Vietnam were found to have 50 percent lower odds of dropping out between the ages of 8 and 12 (Kaffenberger 2021). In contrast, qualitative analysis indicates a direct relationship between low learning and dropout, with children and parents choosing to discontinue school when they realize how little is being learned (Kaffenberger 2021). Lower achievement on academic scores can increase the risk of dropping out by increasing the perceived opportunity cost of sending children to school. In Kenya, among students who dropped out early, falling behind peers and being unable to process the content being taught was cited as the most common reasons for dropping out (Pritchett and Beatty 2015; Zuilkowski et al. 2016). A few studies evaluate how high temperatures will impact long-term education out- comes and human capital production and discuss the potential long-term effects. Park (2022) shows that high temperatures on exam day decreased the likelihood of passing a subject as well as reduced graduation rates for high school students in New York City (Park 2022). Garg et al. (2020) discuss that longer-run temperature can affect agricul- tural income and damage livelihoods, which was found to reduce school attendance (Garg et al. 2020). A study in Cameroon finds a reduction in plantain yields due to cli- mate change which was found to decrease parental investment in education (Fuller et al. 2018). The study predicts that plantain yields will decrease by 39 percent by 2080 due to climate change, which will be associated with a 51 percent reduction in average years of schooling compared to 1991 baseline measures (Fuller et al. 2018). There are many reasons for which dropouts may be a consequence of exposure to high temperatures. In the short term, a hot day could change behavioral patterns in student absenteeism due to potential disturbances in transportation or higher discom- Venegas Marin et al. 199 fort when attending school, particularly for schools that do not have access to air con- ditioning or good air flow. This has been shown for adult labor supply, which decreases with high temperature (Zivin and Shrader 2016). Interestingly, one study even showed study subjects dropping out of the experiment due to thermal discomfort at tempera- tures above 39◦ C (Tian et al. 2021). In the medium term, exposure to high temperatures seems to be associated with lower school attainment. A study in Ethiopia finds that a child who experiences above average spring temperatures (0.5 standard deviations above long-term mean) has 21 percent lower odds of completing one grade of school Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 (Randell and Gray 2019). Tropical Cyclones, Floods, and Education Outcomes Tropical cyclones and flooding can have impacts on education through their effects on learning, attendance, and long-term academic achievement. Learning can be im- pacted through school closures that prevent students from going to school. Even when schools are not closed, issues with transportation to and from the school can hinder attendance and participation, also affecting learning. Extended closures due to infras- tructure damage or the use of schools as evacuation centers have adverse effects on schooling continuity. Lastly, tropical cyclones and floods can also have lasting effects on education through decreased enrollment, increased dropouts, and an overall nega- tive effect on long-term educational achievement. Impacts on Learning Tropical cyclones and storms often close schools, halting all learning. Effects of these storms can include flash floods, landslides, and tidal surges which can have detri- mental impacts on schools and lead to closures. For example, storm Ana in Mozam- bique destroyed almost 800 schools in early 2022, all of which were unable to open fol- lowing the storm (BBC 2022). Another storm, Cyclone Freddy, which affected Malawi, Mozambique, Madagascar, South Africa, and Zimbabwe in March 2023, also affected schools and student attendance; about half a million students in Malawi, or 5 per- cent of students across the county, were not able to attend school due to school clo- sures in the affected areas (Mugo 2023). Another tropical storm in the northern Philip- pines in August 2022 also led to school closures in the capital (AP 2023). Classes are automatically suspended during a typhoon in the Philippines, and it can take over a week to clean the classrooms; this was found to happen at least twice a month in some regions during the rainy season (Cadag et al. 2017). Although school closures are sometimes necessary, they will have consequences on student learning and achieve- ment. There is strong evidence that school closures produce learning losses (Schady et al. 2023). During the first years of the COVID-19 pandemic (March 2020–2022), each month of closures translated to a month of learning losses (Sabarwal et al. 2023). 200 The World Bank Research Observer, vol. 39, no. 2 (2024) We can extrapolate from the drastic impacts of school closures related to COVID-19 that storm and cyclone-driven school closures will also have severe consequences on learning. The vulnerability of school infrastructure increases the risk of extended school clo- sures. The damage extreme weather events cause to school infrastructure will largely determine both the length of the school closure as well as the learning that will be im- pacted. Higher levels of damage mean that schools will take longer to guarantee a safe environment before students can return. In the case of complete school destruction, Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 the provision of education services will take even longer. The extent of the damage ul- timately depends on the school’s resilience, and unfortunately a significant number of schools are unprepared to cope with extreme weather and climate shocks. For exam- ple, the World Bank’s Pacific Safer Schools Program found that between 50 and 90 per- cent of 6,000 school buildings across Samoa, Tonga, and Vanuatu may not withstand a strong cyclone or earthquake (World Bank 2022a). Recent climate shocks highlight similar vulnerabilities in other regions. In Haiti, physical damage to the education sec- tor from natural disasters has damaged one in four schools across the country. The low level of resilience in school infrastructure globally amplifies damages, thereby extend- ing school closures that halt learning. The use of school facilities as evacuation centers can also prolong school closures, affecting education and learning. Schools are often used as such following storms and floods as shown in recent emergencies in the Philippines, Haiti, Libya, Pakistan, and Japan (UNICEF 2016, 2023;Cadag et al. 2017; Perry et al. 2023). Schools used as evacua- tion centers for extended periods not only keep children out of school but can also pro- duce long-term damage to school facilities when improperly managed. In Haiti, Hur- ricane Mathews led to over 100,000 children being out of school due to the facilities being damaged or used as temporary shelters (UNICEF 2016). This was the case even months after the hurricane. In Pakistan, 92 percent of households affected by flooding in 2022 were still uncertain six months later of when local schools would re-open (Perry et al. 2023). The use of schools as evacuation centers can compromise the continuity of schooling accessibility and can have detrimental impacts on enrollment and long- term educational achievement. A study in the Philippines investigated links between school hazards ( floods, earthquakes, and typhoons) and found that the repeated use of school structures for evacuations during hazards had adverse impacts on the aca- demic performance of students (David et al. 2018). After flooding, students also report the destruction of learning materials and the use of schools as evacuation centers as challenges to the learning environment (Cadag et al. 2017). A study in flood-affected regions of Pakistan found that institutions remained closed for over three months due to infrastructure damage or the use of schools as shelter homes for flood victims (Saif- Ur-Rehman and Shaukat 2013). Even when schools are not closed, tropical cyclones, hurricanes, and typhoons can reduce attendance and participation. There are more absences during the rainy season Venegas Marin et al. 201 even when classes are not suspended, often due to challenges in transportation and particularly for more vulnerable students. A study in metropolitan areas of Brazil that evaluated the use of face-to-face and virtual classes for Biology undergraduate students in areas affected by floods found that on flood days, students spent more time travelling from home to the university (2.54h ± 0.52 on flood days vs. 1.24h ± 0.83 on non-flood days) (Santana et al. 2013). There was also a difference in attendance, whereby the per- centage of students present for face-to-face classes decreased from 77 percent on days without flooding to 27 percent on days with flooding (Santana et al. 2013). This influ- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 enced test scores, with those who had only face-to-face classes showing approximately 33 percent lower test scores on average, but no difference was observed when students had access to both face-to-face and virtual learning options. This highlights the poten- tial use of online learning to limit the impacts of flooding on learning in the Brazilian context (Santana et al. 2013). There is specific evidence that the effects of floods and storms adversely affect learn- ing. Several studies have considered how flooding can decrease test scores and the par- ticipation of affected students. A study on climate shocks in Colombia, in which the majority of shocks were floods and landslides, showed negative impacts on test scores in a national standardized exam for high school students (Valencia Amaya 2020). Quali- tative research in Zimbabwe indicated that floods were associated with a loss in learn- ing hours (Mudavanhu 2014). A study from Thailand evaluated the effect of the 2011 floods on test scores and found flooding to be associated with decreases in test scores for most grade levels and subject areas (Thamtanajit 2020). During Hurricane Katrina in New Orleans, communication, issues with housing, and emotional and financial im- pacts were important effects of the hurricane for medical students at Tulane Univer- sity School of Medicine and Louisiana State University School of Medicine (Krane et al. 2007). Typhoons in the Philippines in October and November 2020, concurrent to COVID-19, decreased overall student online participation but participation in assign- ments and quizzes was not affected (Lagmay and Rodrigo 2022). Overall, increasing floods and typhoons will adversely affect student learning and participation, impact- ing student achievement. This can also have long-term effects on schooling and edu- cational outcomes. Impacts on Enrollment, Dropouts, and Long-Term Outcomes Storms such as floods and hurricanes can have detrimental consequences on edu- cational systems that can increase dropouts. Mudavanhu (2014) conducted research in Zimbabwe and found that floods led to a loss of qualified educators, outbreaks of waterborne diseases, higher absenteeism, and lower coverage of learning material, which had long-term effects on education (Mudavanhu 2014). Absenteeism was high- est during the rainy season and attendance was only 50 percent during these months ( January–March). Over 17 percent of children in this region dropped out of school 202 The World Bank Research Observer, vol. 39, no. 2 (2024) due to socioeconomic and behavioral factors that were likely exacerbated by floods (Mudavanhu 2014). After school closures, some students will not return, further increasing dropout rates. School closures also increase dropouts as many students never return to school when it reopens. For example, in Chile, school closures increased the probability of students dropping out of high school by 49–68 percent (Grau et al. 2018). Ev- idence from the COVID-19 pandemic found that school closures were associated with up to 35 percent dropouts; this was higher among students with lower socioe- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 conomic status, potentially exacerbating learning inequalities (Moscoviz and Evans 2022). Following COVID-19 school closures, in Ethiopia and Pakistan, school enroll- ment among children 6–14 dropped by 4 percentage points and 6 percentage points respectively, once schools re-opened post-COVID (Schady et al. 2023). Declines were much larger for students from lower socioeconomic backgrounds. School closures driven by storms/typhoons will show similar effects. For individuals, reduced education attainment will translate into lower earnings and productivity. Dropouts are associated with lower school attainment and will have long- term consequences on income and human capital (McCaul et al. 1992). There is strong evidence that school attainment is linked with higher earnings, with estimates sug- gesting a return of 9–10 percent for each additional year of schooling (Psacharopoulos and Patrinos 2018). The returns are higher specifically among the groups that will be most impacted by climate change—poorer countries and girls. This implies that as cli- mate shocks reduce education attainment, future earnings will also suffer, particularly in poorer countries and among girls. It has been shown that students in grades 1–12 affected by school closures will earn 3 percent less over their lifetimes (Hanushek and Woessmann 2020). It will be important to continue understanding and tracking the long-term effects of hydrometeorological disasters such as floods, hurricanes, and cyclones on education outcomes in the context of climate change. Projections show that Southeast Asia, In- dia, Eastern Africa, and the Andes will be regions particularly affected by flood risk in the context of climate change and should be prioritized in preventative actions to limit destruction (Hirabayashi et al. 2013). Droughts and Education Outcomes Although fewer studies have considered the effects of droughts on learning, enroll- ment, and attendance, there is some evidence that rainfall shocks and droughts can have an impact on cognitive development and education achievement. There are also long-term impacts of droughts on the continuity of education. The evidence of droughts and rainfall shocks on education outcomes shows that they adversely affect learning as well as school attainment. However, negative rainfall shocks are associated Venegas Marin et al. 203 with increased schooling in some contexts. It will be important to better prepare for the consequences of drought-impacted regions on education outcomes. Impacts on Learning Droughts adversely affect learning. A study in the rural areas of the state of Maharash- tra in India found that drought was associated with a 4.14 percent decrease in math scores and 2.67 percent decrease in reading scores for primary and secondary school Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 children (Joshi 2019). A study of the droughts that occurred from October 2015 to April 2016 in Southern Africa found a 4 percent decline in math scores for affected girls in Zimbabwe but literacy scores were not affected (Nordstrom and Cotton 2020). Leader- ship attitudes were also found to be lower for affected students (Nordstrom and Cotton 2020). Shah and Steinberg (2017) found that children in India had a 0.012–0.04 point higher score in math and reading for each additional year with exposure to rainfall early in life in contrast to years with drought and this also increased the probability of being on track with schooling (Shah and Steinberg 2017). Although the evidence base for the effects of drought on learning is not extensive, existing studies have found exposure to droughts to negatively impact learning outcomes. Impacts on Enrollment, Dropouts, and Long-Term Outcomes There are mixed results regarding the effect of droughts on enrollment and long-term educational outcomes. Drought-induced water scarcity can negatively affect school at- tendance and dropouts. A number of studies show a decreased investment in school- ing and enrollment following a drought. The study by Nübler et al. (2021)in pastoralist Kenya found that long-term effects of rainfall shocks led to strain on household re- sources and lower expenditure on schooling, even 10 years after the shock (Nübler et al. 2021). In Yemen, climate change is making rainfall scarcer and less reliable, prompt- ing many children in rural communities to drop out of school to collect limited drink- ing water (World Bank 2022b). Rainfall shocks during early childhood and schooling- age also lowered the probability that the child enrolled in school, and reduced years of schooling by 0.8 (Nübler et al. 2021). Indonesian girls living in areas enriched by rain during childhood are not only healthier and taller but also attain more schooling (Almond et al. 2010). In some contexts, droughts have been found to show a positive effect on attendance. Children are taken out of school to productive work when there is more need; therefore, when a drought occurs the opportunity cost of sending children to school decreases (Shah and Steinberg 2017; Nordstrom and Cotton 2020). In Zimbabwe, compared to those that were unaffected, girls affected by droughts were 2.8 percent more likely to be enrolled in school one year following the drought and 7.9 percent less likely to repeat a grade (Nordstrom and Cotton 2020). Shah and Steinberg (2017) also found that chil- 204 The World Bank Research Observer, vol. 39, no. 2 (2024) dren in India who are affected by droughts are 0.3 percent less likely to drop out the year following the drought and 4 percent more likely to have attended school the previous week (Shah and Steinberg 2017). In Ethiopia, exposure to summer droughts in early childhood decreases odds of completing schooling by 16 percent (Randell and Gray 2019). A study in Colombia also finds that deficit rainfall shocks decrease dropouts and academic performance (Palacios and Rojas-Velásquez 2023). The change in demand for work in agriculture or the household can decrease the cost of sending children to school, having a positive effect on long-term education. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Wildfires and Education Outcomes Wildfires can have impacts on populations and communities affected by the fire it- self through displacement. They can also produce high levels of smoke, affecting much larger areas and populations by spreading far beyond the wildfire perimeters. Smoke from wildfires consists of high levels of fine particulate matter, a pollutant that has detrimental impacts on education outcomes and can also produce school closures in some regions. Recent evidence has also shown air pollution to be associated with lower learning and academic achievement, and several studies have evaluated the specific ef- fect of wildfire smoke. Impacts on Learning Wildfires can cause damage to schools and lead to closures which can have important impacts on learning outcomes. For example, two school buildings collapsed and one was severely damaged during the 2018 Camp Fire in California (Schulze et al. 2020). School closures can also be a response measure from governments and school officials to protect students from wildfire smoke. Wildfire smoke is a form of air pollution, and includes high concentrations of fine particulate matter (PM2.5 ), which has known ad- verse impacts on human health (Reid et al. 2016; Aguilera et al. 2021). School closures can be used to limit student exposure to unhealthy air during transportation to and from school and during the school day. In California, over 31 million student-days of wildfire smoke exposure occurred in the 2020–2021 school year (Velásquez et al. 2023). Wildfires and their smoke have been responsible for the majority of short-term school (1–5 days) closures in California which were shown to adversely affect academic per- formance for younger students, although older students were unaffected (Miller and Hui 2022). During major wildfires in California in 2020, 18 percent of the state’s public school enrollment stayed home due to school closures (Cano 2020). Air pollution adversely affects student learning. There are numerous studies that quantify the impacts of fine particulate matter (PM2.5 ) on academic achievement by comparing changes in exposure to ambient air pollution and test scores in different re- gions around the world including Italy (Bernardi and Keivabu 2023), the United States Venegas Marin et al. 205 (Gilraine and Zheng 2022), the Islamic Republic of Iran (Amanzadeh et al. 2020), Brazil (Carneiro et al. 2021), Israel (Lavy et al. 2014), Chile (Miller and Vela 2013), China (Chen et al. 2017; Zhang et al. 2018), and India (Balakrishnan and Tsaneva 2021). The ma- jority of these studies observe lower test scores for students with higher exposure to PM2.5 , which is a particularly harmful air pollutant due to its small size and ability to be inhaled deeply into the lung alveoli and even bloodstream. A review of the ef- fects of traffic-related air pollution showed that 9 out of 10 studies found poorer aca- demic performance in children and adolescents who were exposed to higher levels of Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 traffic-related air pollution, although the quality of evidence was low (Stenson et al. 2021). There is also evidence that wildfire smoke-specific air pollution impacts student achievement. Wildfire-specific PM2.5 is more harmful to health than PM2.5 from other sources and usually produces exposure at very high concentrations, therefore it is im- portant to evaluate its specific effects (Aguilera et al. 2021). Wen and Burke (2022) used data from standardized test scores in the United States and satellite data to iden- tify wildfire smoke and found that exposure to PM2.5 from wildfire smoke reduces test scores by ∼0.15 percent of a standard deviation (Wen and Burke 2022). This effect is consistent across school subjects and has a higher effect on younger students (Wen and Burke 2022). Wu (2022) conducted a similar analysis and found that exposure to 10 additional wildfire smoke days was associated with a 0.003 standard deviation re- duction in English and 0.004 standard deviation reduction in math scores (Wu 2022). Higher income in the United States has been shown to be much more likely to be able to leave the county during a smoke event, while more disadvantaged populations may not be able to as easily take measures to avoid exposure (Holloway and Rubin 2022). Impacts on Enrollment, Dropouts, and Long-Term Outcomes A few studies have evaluated the long-term outcomes of wildfire smoke exposure and observed substantial educational and economic impacts. Cheng et al. (2023) showed that one standard deviation increase in PM2.5 exposure decreased parental spend- ing in children’s education by up to 44.6 percent (Cheng et al. 2023). Wen and Burke (2022) showed that the effects of wildfire smoke on test scores in one year, 2016, had long-term impacts and reduced future earnings of affected populations in the United States by $1.7 billion, which burdened primarily disadvantaged populations (Wen and Burke 2022). School closures from wildfires and wildfire smoke can affect attendance and enrollment and have effects on children’s educational success (Miller and Hui 2022). These studies highlight the need to continue to study these impacts to under- stand the full extent of how wildfire smoke impacts students and schooling in the long term. 206 The World Bank Research Observer, vol. 39, no. 2 (2024) Indirect Effects of Extreme Weather Events on Education Out- comes Extreme weather events can affect the health, well-being, and social and economic environment of students and children, which will also adversely impact their educa- tional outcomes. Climate extremes have effects on agricultural income and disease prevalence, produce food insecurity and nutrition-related effects, exacerbate mental health conditions, increase infectious diseases, and decrease sleep quality. They can Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 cause harm across the life cycle, starting in utero, which can impact important de- terminants of success in education systems and potentially influence the long-term achievement of students. Extreme weather can also drive migration, conflict, and dis- placement, which can hinder students’ school attendance and academic success. Exposure to climate extremes impacts agriculture, food security, and income which have adverse effects on learning and schooling. Children in developing countries are particularly vulnerable to nutrient deprivation, infectious diseases, and unsafe environments—30 percent of children under 5 in developing countries are physically stunted, typically due to chronic malnutrition (Black et al. 2017). Extreme weather can affect livelihoods, loss of food availability, and even famine (Stanke et al. 2013; Algur et al. 2021). Climate extremes will increase food insecurity, particularly in vulnerable regions. It is estimated that up to 170 million additional people will be at risk of hunger by 2080 due to climate change (Schmidhuber and Tubiello 2007). Poor nutrition and health will have adverse effects on student learning and achievement. In India, agri- cultural income is an important driver of the effect of heat on academic achievement; impacts are stronger during the growing season and for districts without heat-resistant crops (Garg et al. 2020). Income shocks and household coping mechanisms also jeopardize school enroll- ment. In the Republic of Korea, Cho (2017) suggests that the observed effect of hot summer days in decreasing Korean college exam scores is related to a decrease in time to study due to changes in summer job availability or parental income (Cho 2017). In Kenya, families that experienced historical rainfall shocks had about 10 percent less livestock and were less likely to have a metal roof or access to a latrine. This can pro- duce strain on household resources and can lead to lower expenditure on schooling, even 10 years after a shock (Nübler et al. 2021). There are also social responses to cope with these stressors when families are particularly affected by economic strain due to climate. In Bangladesh, exposure to cyclones, floods, and droughts increases child marriages as families use bride payments as a coping mechanism for financial hard- ship (Asadullah et al. 2021). Taking girls out of the classroom to be married early will have drastic implications for their education. The economic strain of climate shocks on households negatively impact learning and access to school. Droughts can influence the likelihood of parents sending their kids to school, and lower teacher attendance Venegas Marin et al. 207 due to decreases in wages, which are all potential mechanisms in affecting schooling (Shah and Steinberg 2017). There are actions that families may take in response to the economic strain caused by extreme weather events that will be highly impactful for student’s well-being and education. Exposure to extreme weather events has also been shown to impact mental health symptoms and conditions, which adversely affects academic achievement. Heat stress can produce mood disorders, anxiety, and other mental health conditions (Berry et al. 2010; Cianconi et al. 2020). A literature review on the associations between tempera- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 ture and mental health outcomes also shows a higher risk of suicide from heat across studies and a higher risk of hospital admissions and emergency department visits for mental health (Thompson et al. 2018). Droughts can increase stress and are associated with increased suicides for farmers (Padhy et al. 2015). A study with 94 ninth grade eth- nic minority students exposed to Hurricane Katrina found that the majority of students (69 percent) had mild-to-severe symptoms of post-traumatic stress disorder (Weems et al. 2009). Children and adolescents who experienced a wildfire disaster also have a higher risk of mental health symptoms, including depression and posttraumatic stress disorder (Brown et al. 2021; Ritchie et al. 2021), lower perceived quality of life, and lower self-esteem (Brown et al. 2019). A study evaluating the mental health of college students after the Fort McMurray wildfires showed that the prevalence of posttraumatic stress disorder in the study population increased by 11.0 percent after the wildfire (Ritchie et al. 2021). Another study that surveyed students in grades 7-to-12 for 3.5 years follow- ing the Fort McMurray wildfires showed that mental health symptoms worsened with time following the disaster and certain groups such as females, older students, and those with a minority gender identity were particularly susceptible (Brown et al. 2021). Climate anxiety can also be an important stressor for youth (Crandon et al. 2022). Poor mental health can prevent students from having good study habits, increase procrasti- nation and can lower achievement in school. College students with major depression have poorer study habits, greater procrastination, and lower academic performance (Jeffries and Salzer 2022). Changes in climate such as higher temperatures and precipitation can also affect the distribution and spread of certain infectious diseases that adversely influence chil- dren’s schooling attendance and educational achievement. Vector-borne diseases such as malaria, dengue and Lyme disease are highly sensitive to climate and are expected to increase in many global regions in the context of climate change (Caminade et al. 2019). Additionally, the suitable conditions for some of these vectors are shifting to new locations, producing a burden in regions and populations that were previously unex- posed and have decreased immunity to these diseases (Mordecai et al. 2020). Children in Uganda have lower reading scores after cerebral malaria, a type of neurological com- plication from the malaria parasite most prevalent in Africa (Nakitende et al. 2023). A study in Denmark found that students that were hospitalized for an infection in child- 208 The World Bank Research Observer, vol. 39, no. 2 (2024) hood had 0.82 (95 percent confidence interval: 0.79–0.85) lower odds of completing ninth grade (Köhler-Forsberg et al. 2018). Extreme weather events can affect sleep, which is an important driver of academic performance and achievement (Eide and Showalter 2012). Higher temperature has been associated with increases in sleep apnea and other sleep problems (Rifkin et al. 2018). A study in the United States found that when monthly nighttime temperature changes by 1◦ C, a three-night increase in insufficient sleep is observed per 100 per- sons (Obradovich et al. 2017). Sleep disturbance is also an effect of wildfires—a sys- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 tematic review of the impact of wildfires on sleep found that approximately 63–72.5 percent of wildfire survivors have insomnia and 33.3–46.5 percent have nightmares up to 10 months following the disaster (Isaac et al. 2021). A review on climate and sleep by Rikfin et al. (2018) found a higher rate of disrupted sleep for those affected by the 2011 summer floods in Brisbane, Australia and flood victims in China which was re- lated to fear and depression (Rifkin et al. 2018). Another study on sleep problems and posttraumatic stress symptoms for elementary school-aged children exposed to Hurri- cane Ike in Texas found that sleep problems were prevalent and that they persisted in the second year following the disaster (Lai et al. 2020). The optimal number of hours of sleep is higher for younger populations and lower sleep is associated with lower scores in math and reading tests (Eide and Showalter 2012). For example, students at age 12 and 16 perform 0.035 and 0.045 standard deviations lower on tests, respectively, when sleeping 1 hour less than their optimal hours (Eide and Showalter 2012). Learning and education is also impacted indirectly through health shocks that affect children starting in-utero. A study on climate change and education in tropical regions found that a child in Southeast Asia exposed to high temperatures (2 SDs above aver- age) in the prenatal period or early in life has 1.5 less years of schooling than a child with average temperature exposure (Randell and Gray 2019). Rainfall shocks during a child’s first year of life in a pastoralist region of Kenya lowered academic achievement and cognitive skills, reducing reading skills by 13 percent for example, although shocks that occurred later in life were found to be less impactful (Nübler et al. 2021). A study evaluating the effect of El Niño droughts found that children who were not exposed to extreme weather conditions in utero and early childhood had a 5 percent higher likelihood of staying in school and an increase of 0.05 standard deviation in scores for high school exit exams (Duque et al. 2018). Another study found that disability rates in- creased by 3.5–5.2 percent for those exposed to droughts in infancy for South Africans who were sent to live in rural and isolated pockets of land during apartheid (Dinkelman 2017). Exposure to extreme heat in utero has also been associated with increased risk of preterm birth and low birth weight (Bekkar et al. 2020). Preterm infants have been shown to have 1.57 (1.33–1.86) higher odds of having special education needs in school (Odd et al. 2016) and are at higher risk of poor educational achievement (Quigley et al. 2012). A review of the executive functions of children born preterm or low birthweight found that they had lower working memory (by 0.5 of a standardized mean) and inhi- Venegas Marin et al. 209 bition (by 0.4 of a standardized mean) (van Houdt et al. 2019). This will produce inter- generational effects from extreme weather events by impacting the future schooling and education of children even before birth. In the long term, high temperatures and extreme weather events can drive conflict, migration, and displacement that can affect the ability of students to have continuous learning and school attendance. It is projected that by 2050, tens of millions of peo- ple will be forced to migrate within their countries due to climate change (Rigaud et al. 2018). A meta-analysis on climate and conflict found that a one standard deviation Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 change in climate (temperature and rainfall) can increase the risk of intergroup con- flict by 14 percent and of interpersonal violence by 4 percent (Hsiang et al. 2013). Con- flict, violence, and war have consequences for children’s educational attainment and achievement. A study from Burundi not only found that children exposed to war were stunted but that they completed 0.7 fewer years of school, on average, and later in life earned 21 percent less (Akresh 2016). Similarly, a study from Peru found that children exposed to Peru’s 1980–93 civil war early in life had 0.3 less years of schooling in adult- hood (Galdo 2013). A study of populations affected by the Bosnian war also found that displaced parents spend less on their children’s education (Eder 2014). Evidence from the study on the Peruvian civil war not only showed that children exposed to the war in the first three years of their lives had lower education but that this translated to 5 percent lower monthly earnings in adulthood and a lower probability of working in the formal economy (Galdo 2013). High temperatures can also have long-term impacts on national wealth (Zivin and Shrader 2016). Ultimately, the pathways through which ex- treme weather impact education outcomes unequivocally result in lower learning and educational attainment, both of which are key to reap the economic benefits of edu- cation. As we witnessed with the COVID-19 pandemic, resulting learning losses and lower levels of educational attainment reduce incomes, productivity, and amplify ex- isting inequalities (Azevedo et al. 2022). Protecting and Leveraging the Education Sector To minimize detrimental impacts of climate change on education outcomes, it will be important to promote adaptation and resilience in the education sector. This is partic- ularly urgent because these detrimental impacts will continue to become more severe. However, the scale of the impact will depend on the ultimate depth of the climate crisis. A country like Ethiopia will experience a median of 138 hot days (above 35◦ C) under a pessimistic (SSP5-8.5) scenario, compared to 81 days under a middle of the road (SSP2- 4.5) scenario; implying very different impacts on education outcomes. While adapta- tion is undeniably needed to meet the challenges of already-happening climate change, mitigation can still play a critical role in limiting the magnitude of the impacts. The edu- cation sector is uniquely positioned to galvanized climate action (fig. 2). Education can spur mitigation through changing mindsets and behavior, producing skills required for 210 The World Bank Research Observer, vol. 39, no. 2 (2024) Figure 2. The Bi-Directional Relationship of Climate Change and Education Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Source: Author’s illustration. a green transition and prompting innovative strategies and action for climate mitiga- tion and adaptation. The Need to Urgently Adapt Education Systems Even if the most drastic climate mitigation strategies were implemented, we will con- tinue to observe increasing frequency and intensity of extreme weather events which have detrimental impacts on education outcomes. Actions can be implemented to increase the capacity of education systems to adapt and cope with these increas- ingly prevalent climate stressors, such as encouraging investments in adaptation (Behrer 2023). Prior to the construction of new schools, an evaluation of predicted future climate impacts including sea level rise, ambient temperatures and extreme weather events can be used to select appropriate locations (Theirworld 2018). In ex- isting schools, integrating climate-resilient strategies will be increasingly critical. In Venegas Marin et al. 211 response to extreme heat, schools and classrooms can ensure ventilation and/or air conditioning to keep temperatures at a comfortable level for students to ensure the best learning outcomes. For cyclones, schools in at-risk regions can implement mea- sures to prevent infrastructure damage and limit school closures during a storm and their associated impacts. For droughts, school nutrition programs can be effective to prevent impacts on food security and associated learning. For wildfire smoke, air filtra- tion systems can be installed in schools and classrooms to reduce student exposure to air pollution. Furthermore, actions can be implemented to predict, prepare and assess Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 extreme weather events and incorporate disaster response in education sector plans. More than 40 percent of World Bank investments in the education sector are focused on education facilities, creating an opportunity to increase their resilience to climate stressors (World Bank 2024). Interventions can be implemented at the school and classroom level to reduce the burden of extreme heat for students in the classroom. Strategies to include thermal comfort in schools include increasing natural ventilation and/or accessibility to air conditioning in the classroom (Haverinen-Shaughnessy and Shaughnessy 2015; Singh et al. 2019). A meta-analysis on the relationship between classroom temperature and children’s performance estimated that reducing classroom temperature from 30◦ C to 20◦ C could increase performance in learning-related tasks by 20 percent (Wargocki et al. 2019). Porras-Salazar et al. (2018) conducted a cross-over trial with elementary school students in Costa Rica in which air conditioning units were used to reduce classroom temperatures from about 30 to 25◦ C. The 11-year-old students showed an improved performance in lower classroom temperatures; speed in cognitive tests in- creased up to 7.5 percent, and accuracy increased by 0.6 percent for each degree of reduction in classroom temperature (Porras-Salazar et al. 2018). Interestingly, this ef- fect was stronger for lower performing students. The authors suggest that decreasing indoor temperatures can be a strategy to combat inequalities since the children who performed worse on exams were the most affected by high classroom temperatures (Porras-Salazar et al. 2018). Resilience-building initiatives that target root causes focusing on disadvantaged schools and populations can also be effective strategies to limit detrimental impacts of heat (Sheffield et al. 2017). Ensuring natural ventilation and climate-responsive designs of schools can be alternative strategies to air conditioning which may not be feasible in all contexts. For existing schools, less costly options range from painting rooftops with solar-reflective white paint, increasing tree coverage in and around the school, and even adapting school schedules to avoid hours with extreme heat. For new school constructions, the use of certain materials such as clay to offer thermal protection or the design of roofs to ensure ventilation can be effective in limiting heat in classrooms (see Gando Primary School as a good example). School construction integrating natu- ral daylight and cross-ventilation as well as trees and/or shade structures can reduce the energy needed (Chalupka and Anderko 2019). For example, Kenya implemented a 212 The World Bank Research Observer, vol. 39, no. 2 (2024) Green Economy Strategy and Implementation Plan that promotes bioclimatic design for school buildings and will increase thermal comfort for students during high tem- peratures (Odera 2020). Adapting education systems to be more resilient to typhoons, droughts and wildfire smoke may involve action at multiple levels, including increasing the resilience of the school as well as establishing learning recovery or nutrition programs. Increasing the resilience of the school to prevent closures during disaster will likely be the most effec- tive strategy, and can include actions such as construction of walls to prevent damage Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 from floods or changes in land-use management (Theirworld 2018). Schools and ed- ucation systems can also implement actions to limit their effects on student achieve- ment and attendance. There are some good examples of addressing learning losses after school closures. An after-school program in Tamil Nadu, India helped recover half of the learning losses from COVID-related school closures via 60–90 minutes of daily remedial learning for six months (Singh et al. 2022). School nutrition programs can be effective in reduc- ing impacts of droughts on nutritional outcomes (Taras 2005). Installing air filtration systems has been shown to be an effective adaptation strategy to reduce exposure to wildfire smoke (May et al. 2021). One study evaluated potential benefits of installing air filtration systems for student achievement and found that it can increase student achievement and performance (Gilraine 2023). Forecasting and predicting extreme weather events can be important not only to prevent adverse health impacts of these events, but also their impacts on education outcomes. If schools receive sufficient warning about a potential climate shock, they may be able to prepare and protect the students that are most vulnerable. Novel forecast-based financing systems have also been proposed to identify and provide funding support to regions for post-disaster response (Coughlan de Perez et al. 2015). By identifying at-risk regions before the climate shock occurs, appropriate actions can be activated to increase resilience. Nearly 60 percent of countries in a 2017 survey of 68 high-risk countries for disasters include either disaster risk reduction or disaster response components in their education sector plan, but these are not always compre- hensive (GPE 2023). For example, the Philippines implemented early warning systems for typhoons, floods, and earthquakes through a mobile app that is used by the edu- cation system to improve the disaster knowledge of students and staff and help them take action (Aranda 2022). Policymakers and educational planners should work together to ensure climate- ready schools and systems in the context of climate change and the link between disas- ter planning frameworks and the education sector should be strengthened (GPE 2023). The involvement of students in disaster risk reduction and management at local, re- gional, and global levels is an underutilized strategy that can be critical for resilience development (Pal et al. 2023). Venegas Marin et al. 213 It has been estimated that the increased mortality risk from temperature change corresponds to approximately 3.2 percent of global GDP by 2100 under a high emis- sions scenario, which will be worse in poorer regions of the world (Carleton et al. 2022). While adaptation measures will be essential to limit destructive impacts, they do have significant costs that will continue to increase in the context of climate change. For example, per capita median annual adaptation costs in developing countries are pro- jected anywhere from US$17 to US$26 in 2050, depending on which climate path we take (Chapagain et al. 2020). Also, adaptation strategies will be increasingly costly. As Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 extreme weather events become increasingly severe, the effectiveness of these mea- sures decreases due to the higher frequency and severity of these events. Therefore, re- lying solely on adaptation will not be effective in order to avoid the detrimental impacts of climate change, and there is copious evidence to show the importance of acting on climate now to prevent catastrophic impacts. The Need to Leverage the Education Sector for Climate Action Empowering children and youth with education paves the way for effective climate change adaptation and mitigation. There is a strong and positive relationship between education and the capacity to adapt to climate change. More educated individuals are better able to prepare for, cope with, and recover from extreme weather events. Evidence from Senegal, Mali, Thailand, Cuba, Haiti, the Dominican Republic, El Salvador, and Brazil provides robust evidence for the posi- tive impact of education on vulnerability reduction (Wamsler et al. 2012; Muttarak and Pothisiri 2013; Pichler and Striessnig 2013; Van der Land and Hummel 2013). In these studies, people with higher levels of education exhibit greater disaster prepared- ness and response, experience reduced adverse effects, and recover more quickly from disasters. Evidence suggests that there are direct and indirect pathways that support this pos- itive relationship. Education attainment directly influences risk perception, skills, and knowledge, all of which empower individuals to be better prepared against extreme weather events and thus reduce impacts. Education, from early years to adulthood, promotes cognitive development and vital skills that are key for adaptive capacity. UNICEF estimates that improving educational outcomes could reduce the climate risks borne by 275 million children globally (UNICEF 2021b). Indirectly, higher lev- els of educational attainment reduce poverty, improve health, and slow down popu- lation growth, all of which are linked with higher community-level adaptive capacity (Muttarak and Lutz 2014). Beyond adaptation, evidence also suggests a strong and positive relationship be- tween educational attainment and mitigation practices. An additional year of educa- tion has been found to increase pro-climate beliefs by 6.3 percent, change behavior by 8.5 percent, and produce a 35 percent increase in green voting across 16 European 214 The World Bank Research Observer, vol. 39, no. 2 (2024) countries (Angrist et al. 2023). In China, educational attainment is associated with a 2 percent increase in pro-environmental attitudes and behaviors (Wang et al. 2022). Similarly, in Thailand, a study found that additional years of schooling are associated with knowledge-based environmentally friendly actions such as increasing regular use of cloth bags by 5 percent and energy-efficient appliances by 7.7 percent (Chankrajang and Muttarak 2017). Globally, the level of education attained emerges as the most influ- ential factor in predicting climate change awareness, with an understanding of human- induced climate change being the primary predictor of perceptions regarding climate Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 change risks (Lee et al. 2015). Education also exhibits a robust correlation with environ- mental concern and support for policies that benefit the environment (Chankrajang and Muttarak 2017). Education can help galvanize mindset and behavior shifts through exposure to timely information, ideas, and nudges to action. As much as 20–37 percent of the emis- sions reduction needed to achieve net zero depends on individual and household be- havior change (Williamson et al. 2018). Students who completed a year-long university course focused on topics related to climate change managed to lower their personal carbon emissions by an annual total of 2.86 tons of CO2 (Cordero et al. 2020). Address- ing climate change also requires demand for policy changes that prioritize climate ac- tion and willingness to bear the costs of those changes. Studies show that there is a strong positive correlation between education, concern for the environment, and be- haviors that support political decisions that have a positive impact on the environment (Franzen and Vogl 2013; Meyer 2015; Chankrajang and Muttarak 2017). The education sector is also uniquely positioned to develop the skills needed to advance the green transition. The shift of the global economy towards more sustain- able and climate-resilient growth that prioritizes low-carbon development precipi- tates the need for a skilled workforce that can carry out the array of tasks associated with greener jobs. Studies show that there is a strong positive correlation between ed- ucation and the kinds of skills (e.g., problem-solving and critical thinking) that support policies that have a positive impact on climate change (Franzen and Vogl 2013; Meyer 2015; Chankrajang and Muttarak 2017). Compared to non-green jobs, green occupa- tions exhibit a stronger intensity of high-level cognitive skills (Consoli et al. 2016). Ed- ucation can play a critical role in building skills to power the transition to green and resilient economies and jobs. Finally, as the quest for innovative solutions to bolster climate adaptation and mit- igation gains momentum, educational institutions have the potential to thrive as dy- namic centers of innovation. In terms of mitigation, innovation is key to eventually achieving net-zero emissions. Almost 50 percent of the emission reductions to achieve net-zero by 2050 depend on technologies that are at the prototype or development stage (IEA 2020). Even in the case of diffusing existing technologies, contextualizing and scaling up these technologies requires significant research to find effective ways of enhancing deployment. Investing in innovative technologies is associated with lower Venegas Marin et al. 215 CO2 across emerging countries (Afrifa et al. 2020). Universities can be hubs of innova- tion and knowledge creation to innovate and advocate for solutions to climate change (Leal Filho et al. 2023).5 Conclusion Evidence unequivocally shows that extreme weather events—high temperatures, trop- Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 ical cyclones, droughts, floods, and wildfires—harm learning, educational attainment, and human capital accumulation. The harm comes in the form of direct and indirect impacts, especially among the most vulnerable children in the most vulnerable set- tings. As climate change increases the frequency and intensity of extreme weather events, these impacts and the inequities they amplify are likely to grow. In the absence of effective adaptation measures, with more frequent school closures due to extreme weather events and less comfortable classrooms due to high temperatures, learning outcomes and enrollment rates will be significantly and negatively impacted. Thus, eroding a lot of the progress made on education outcomes in recent decades. Both mitigation and adaptation are urgently needed. Climate resilience in schools can be increased by infrastructure and programmatic changes. Improvements in in- frastructure such as access to air conditioning, low-cost strategies to control classroom temperatures, air filtration systems and walls to prevent flooding can protect students from the adverse impacts of heat, wildfire smoke, and floods. Providing nutrition and remedial learning programs can also be effective strategies to increase the resilience of affected students. The costs of adaptation will become increasingly severe under cli- mate change, and our ability to implement climate change strategies will determine the extent of the need for adaptation. Climate change mitigation often outweighs the costs and can have benefits on other sectors. Education can play a catalytic role in climate change mitigation and adaptation by reshaping mindsets, behaviors, voting preferences, skills and innovation. In general terms, educational attainment is strongly correlated with higher adaptive capacity for individuals and their communities. It is also found to be intrinsically associated with preferences and behaviors that support climate mitigation. Emission reductions will need to come from individual behavior change alongside larger national plans to shift economic activity away from carbon-intensive growth and towards greener and more sustainable growth models. These large-scale shifts, largely captured in Nationally De- termined Contributions (NDCs), precipitate the need for a population that will support these policies and a skilled workforce capable of implementing them. In this context, the education sector is uniquely positioned to empower children, youth, and their com- munities with the mindsets, behaviors, skills, and innovative ideas that this climate cri- sis requires. In the words of Mere Vadai, a youth climate activist in Fiji, “The classroom is where we begin transforming climate knowledge into action.” 216 The World Bank Research Observer, vol. 39, no. 2 (2024) Data availability Regarding data access, no new data were generated or analyzed during this study. Conflict of Interest The authors declare no conflicts of interest. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Notes Sergio Venegas Marin (corresponding author), email: svenegasmarin@worldbank.org ; Lara Schwarz, email: lnschwar@health.ucsd.edu; Shwetlena Sabarwal, email: ssabarwal@worldbank.org 1. First three pages of Google Scholar search of search terms climate AND impact AND economic, in- cluding only articles published 2010 and onwards that are review articles on the broad impacts on econ- omy/society and excluding articles on a specific sector, with a focus on methods, or that have a specific regional focus. 2. A Web of Science search on the topic “climate and impact” with the search terms “health” resulted in 24,980, “economic” produced 31,243, and “education” produced 5,732. From these results, we can conclude that there is four-times more research on the economic impacts of climate and five-times more research on the health impacts of climate than there is research considering the educational impacts of climate. Out of the 5,732 results from Web of Science on climate impacts on education, 1,903 (33 percent) are based in the United States and 4,467 (78 percent) are from high-income economies (based on World Bank clas- sifications). This is based on Web of Science classifications and may not describe where the research for the manuscript is based. 3. To estimate days of learning lost, it was assumed that students are expected to learn about 0.3 stan- dard deviations in a given academic year (Schady et al. 2023), and such a year is comprised of about 180 school days. Days of learning lost were estimated with the following equation: Days of learning lost = SDchange 0.3 LAYS ∗9 months∗20 days 4. When estimating overall effects of classroom temperature on test scores, studies ranged from elementary- to college/university-level students. Estimates assumed that the effect of temperature on achievement is linear. 5. A comprehensive analysis of how education can support climate change mitigation and adaptation can be found in Sabarwal et al. ( forthcoming). References Afrifa, G. A., I. Tingbani, F. Yamoah, and G. Appiah. 2020. “Innovation Input, Governance and Climate Change: Evidence from Emerging Countries.” Technological Forecasting and Social Change 161: 120256. Aguilera, R., T. Corringham, A. Gershunov, and T. Benmarhnia. 2021. “Wildfire Smoke Impacts Respiratory Health More than Fine Particles from Other Sources: Observational Evidence from Southern Califor- nia.” Nature Communications 12: 1493. Akresh, R. 2016. “Climate Change, Conflict, and Children.” The Future of Children 26(1): 51–71. Algur, K. D., S. K. Patel, and S. Chauhan. 2021. “The Impact of Drought on the Health and Livelihoods of Women and Children in India: A Systematic Review.” Children and Youth Services Review 122: 105909. Almond, D., L. Edlund, H. Li, and J. Zhang. 2010. “Long-Term Effects of Early-Life Development: Evidence from the 1959 to 1961 China Famine.” In The Economic Consequences of Demographic Change in East Asia, edited by T. Ito and A. Rose, 321–45. Chicago: University of Chicago Press. Venegas Marin et al. 217 Amanzadeh, N., M. Vesal, and S. F. F. Ardestani. 2020. “The Impact of Short-term Exposure to Ambient Air Pollution on Test Scores in Iran.” Population and Environment 41: 253–85. Angrist, N., K. Winseck, H. A. Patrinos, and J. S. G. Zivin. 2023. “Human Capital and Climate Change.” NBER Working Paper No. 31000, National Bureau of Economic Research, Cambridge, MA. AP. 2023. “Storm Forces School Closures, Evacuations in Philippines.” AP News, August 23. https://apnews.c om/article/storms- philippines- manila-evacuations-tropical-1442569fc1d5eba3c593273b8eeaab49. Aranda, C. H., and E. Humeau. 2022. “Early Warning Systems in the Philippines: Building Resilience through Mobile and Digital Technologies.” London, UK: GSMA. Asadullah, M. N., K. M. M. Islam, and Z. Wahhaj. 2021. “Child Marriage, Climate Vulnerability and Natural Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Disasters in Coastal Bangladesh.” Journal of Biosocial Science 53: 948–67. Azevedo, J. P., M. Akmal, M. - H. Cloutier, H. Rogers, and Y. N. Wong. 2022. “Learning Losses During Covid- 19.” Policy Research Working Paper No 10218, World Bank, Washington DC. Balakrishnan, U., and M. Tsaneva. 2021. “Air Pollution and Academic Performance: Evidence from India.” World Development 146: 105553. Barbic, F., M. Minonzio, B. Cairo, D. Shiffer, L. Cerina, P. Verzeletti, F. Badilini, M. Vaglio, A. Porta, M. Sa- tambrogio, R. Gatti, S. Rigo, A. Bisoglio, and R Furlan. 2022. “Effects of a Cool Classroom Microclimate on Cardiac Autonomic Control and Cognitive Performances in Undergraduate Students.” Science of the Total Environment 808: 152005. Barbic, F., M. Minonzio, B. Cairo, D. Shiffer, A. Dipasquale, L. Cerina, A. Vatteroni, V. Urechie, P. Verzeletti, F. Badilini, M. Vaglio, R. Latrino, A. Porta, M. Santambrogio, R. Gatti, and R Furlan. 2019. “Effects of Different Classroom Temperatures on Cardiac Autonomic Control and Cognitive Performances in Un- dergraduate Students.” Physiological Measurement 40: 054005. Bashmakov, I., L. Nilsson, A. Acquaye, C. Bataille, J. Cullen, S. de la Rue du Can, M. Fischedick, Y. Geng, and K. Tanaka 2022. “Chapter 11.” In Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change, edited by N. Campbell and R. Pichs-Madruga. Berkeley, CA: Lawrence Berkeley National Lab.(LBNL). BBC. 2022. “Mozambique’s Storm Ana Aftermath: 778 Schools Destroyed.” BBC, February 1. https://www. bbc.com/news/world-africa-60219693. Behrer, A. P. H., Alaka. 2023. “Education and Climate Change: The Critical Role of Adaptation Investments.” World Bank (blog), February 2. https://blogs.worldbank.org/developmenttalk/education- and- climate- change- critical- role- adaptation-investments. Bekkar, B., S. Pacheco, R. Basu, and N. DeNicola. 2020. “Association of Air Pollution and Heat Exposure with Preterm Birth, Low Birth Weight, and Stillbirth in the US: A Systematic Review .” JAMA Network Open 3: e208243–. Bernardi, F., and R. C. Keivabu. 2023. “Poor Air at School and Educational Inequalities by Family Socioeco- nomic Status.” Rostock, Germany: Max Planck Institute for Demographic Research. Berry, H. L., K. Bowen, and T. Kjellstrom. 2010. “Climate Change and Mental Health: A Causal Pathways Framework.” International Journal of Public Health 55: 123–32. Black, M. M., S. P. Walker, L. C. Fernald, C. T. Andersen, A. M. DiGirolamo, C. Lu, D.C. McCoy, G. Fink, Y.R. Shawar, J. Shiffman, A.E. Devercelli, Q.T. Wodon, E. Vargas-Baron, and S. Frantham-McGregor 2017. “Early Childhood Development Coming of Age: Science through the Life Course.” Lancet 389: 77–90. Brink, H. W., M. G. Loomans, M. P. Mobach, and H. S. Kort. 2021. “Classrooms’ Indoor Environmental Condi- tions Affecting the Academic Achievement of Students and Teachers in Higher Education: A Systematic Literature Review.” Indoor Air 31: 405–25. Brown, M. R., V. Agyapong, A. J. Greenshaw, I. Cribben, P. Brett-MacLean, J. Drolet, C.B. McDonald-Harker, J. Omeje, B. Lee, M. Mankowski, S. Noble, D.T. Kitching, and P. Silverstone 2019. “After the Fort McMurray Wildfire There Are Significant Increases in Mental Health Symptoms in Grade 7–12 Students Compared to Controls.” BMC Psychiatry 19: 1–11. 218 The World Bank Research Observer, vol. 39, no. 2 (2024) Brown, M. R., H. Pazderka, V. I. Agyapong, A. J. Greenshaw, I. Cribben, P. Brett-MacLean, J. Drolet, C.B. McDonald-Harker, J. Omeje, B. Lee, M. Mankowski, S. Noble, D.T. Kitching, and P. Silverstone 2021. “Mental Health Symptoms Unexpectedly Increased in Students Aged 11–19 Years during the 3.5 Years after the 2016 Fort McMurray Wildfire: Findings from 9,376 Survey Responses.” Frontiers in Psychiatry 12: 676256. Cadag, J. R. D., M. Petal, E. Luna, J. Gaillard, L. Pambid, and G. V. Santos. 2017. “Hidden Disasters: Recurrent Flooding Impacts on Educational Continuity in the Philippines.” International Journal of Disaster Risk Reduction 25: 72–81. Caminade, C., K. M. McIntyre, and A. E. Jones. 2019. “Impact of Recent and Future Climate Change on Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Vector-Borne Diseases.” Annals of the New York Academy of Sciences 1436: 157–73. Cano, R. 2020. “School Closures from California Wildfires this Week Have Kept More than a Million Kids Home.” CalMatters, November 15. https://calmatters.org/environment/2018/11/school-closures-calif ornia- wildfires- 1- million- students/. Carleton, T., A. Jina, M. Delgado, M. Greenstone, T. Houser, S. Hsiang, A. Hultgren, R.E. Kopp, K.E. McCusker, I. Nath, J. Rising, A. Rode, H. Kwon Seo, A. Viaene, J. Yuan, and A. Tianbo Zhang 2022. “Valuing the Global Mortality Consequences of Climate Change Accounting for Adaptation Costs and Benefits.” Quarterly Journal of Economics 137: 2037–105. Carneiro, J., M. A. Cole, and E. Strobl. 2021. “The Effects of Air Pollution on Students’ Cognitive Perfor- mance: Evidence from Brazilian University Entrance Tests.” Journal of the Association of Environmental and Resource Economists 8: 1051–77. Caruso, G., I. de Marcos, and I. Noy 2024. "Climate changes affect human capital" Economics of Disasters and Climate Change, 8:157–196 Chalupka, S., and L. Anderko. 2019. “Climate Change and Schools: Implications for Children’s Health and Safety.” Creative Nursing 25: 249–57. Chankrajang, T., and R. Muttarak. 2017. “Green Returns to Education: Does Schooling Contribute to Pro- Environmental Behaviours? Evidence from Thailand.” Ecological Economics 131: 434–48. Chapagain, D., F. Baarsch, M. Schaeffer, and S. D’haen. 2020. “Climate Change Adaptation Costs in Devel- oping Countries: Insights from Existing Estimates.” Climate and Development 12: 934–42. Chen, X., X. Zhang, and X. Zhang. 2017. “Smog in Our Brains: Gender Differences in the Impact of Exposure to Air Pollution on Cognitive Performance”. IFPRI Discussion Paper 1619. International Food Policy Research Institute (IFPRI), Washington, D.C. Cheng, Z., L. Guo, M. Tani, and S. Cook. 2023. “Air Pollution and Education Investment”. IZA Discussion Papers 16322. Institute of Labor Economics (IZA). Cho, H. 2017. “The Effects of Summer Heat on Academic Achievement: A Cohort Analysis.” Journal of En- vironmental Economics and Management 83: 185–96. Cianconi, P., S. Betrò, and L. Janiri. 2020. “The Impact of Climate Change on Mental Health: A Systematic Descriptive Review.” Frontiers in Psychiatry 11: 74. Consoli, D., G. Marin, A. Marzucchi, and F. Vona. 2016. “Do Green Jobs Differ from Non-Green Jobs in Terms of Skills and Human Capital? Research Policy 45: 1046–60. Cordero, E. C., D. Centeno, and A. M. Todd. 2020. “The Role of Climate Change Education on Individual Lifetime Carbon Emissions.” PLoS ONE 15: e0206266. Coughlan de Perez, E., B. van den Hurk, M. Van Aalst, B. Jongman, T. Klose, and P. Suarez. 2015. “Forecast- Based Financing: An Approach for Catalyzing Humanitarian Action Based on Extreme Weather and Climate Forecasts.” Natural Hazards and Earth System Sciences 15: 895–904. Crandon, T. J., J. G. Scott, F. J. Charlson, and H. J. Thomas. 2022. “A Social-Ecological Perspective on Climate Anxiety in Children and Adolescents.” Nature Climate Change 12: 123–31. Das Gupta, M. 2014. “Population, Poverty, and Climate Change.” World Bank Research Observer 29: 83–108. Venegas Marin et al. 219 David, C. C., S. L. C. Monterola, A. Paguirigan, Jr, E. F. T. Legara, A. B. Tarun, R. C. Batac, and J.P. Osorio 2018. “School Hazard Vulnerability and Student Learning.” International Journal of Educational Research 92: 20–29. Davies, P., and I. Maconochie. 2009. “The Relationship between Body Temperature, Heart Rate and Respi- ratory Rate in Children.” Emergency Medicine Journal 26: 641–3. Davis, C. R., S. R. Cannon, and S. C. Fuller. 2021. “The Storm after the Storm: The Long-Term Lingering Impacts of Hurricanes on Schools. Disaster Prevention and Management.” An International Journal 30: 264–78. Dinkelman, T. 2017. “Long-Run Health Repercussions of Drought Shocks: Evidence from South African Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Homelands.” Economic Journal 127: 1906–39. Duque, V., M. Rosales-Rueda, and F. Sanchez. 2018. “How Do Early-Life Shocks Interact with Subsequent Human-Capital Investments? Evidence from Administrative Data.” Economics Working Paper Series 2019-17, Sydney, Australia. Ebi, K. L., J. Vanos, J. W. Baldwin, J. E. Bell, D. M. Hondula, N. A. Errett, K. Hayes, C.E. Reid, S. Saha, J. Spec- tor, and P. Berry 2021. “Extreme Weather and Climate Change: Population Health and Health System Implications.” Annual Review of Public Health 42: 293–315. Eder, C. 2014. “Displacement and Education of the Next Generation: Evidence from Bosnia and Herzegov- ina.” IZA Journal of Labor & Development 3: 1–24. Eide, E. R., and M. H. Showalter. 2012. “Sleep and Student Achievement.” Eastern Economic Journal 38: 512–24. Franzen, A., and D. Vogl. 2013. “Two Decades of Measuring Environmental Attitudes: A Comparative Anal- ysis of 33 Countries.” Global Environmental Change 23: 1001–8. Fuller, T. L., P. R. S. Clee, K. Y. Njabo, A. Tróchez, K. Morgan, D. B. Meñe, N.M. Anthony, M.K Gonder, W.R. Allen, R. Hanna, and T.B. Smith 2018. “Climate Warming Causes Declines in Crop Yields and Lowers School Attendance Rates in Central Africa.” Science of the Total Environment 610: 503–10. Galdo, J. 2013. “The Long-Run Labor-Market Consequences of Civil War: Evidence from the Shining Path in Peru.” Economic Development and Cultural Change 61: 789–823. Garg, T., M. Jagnani, and V. Taraz. 2020. “Temperature and Human Capital in India.” Journal of the Associ- ation of Environmental and Resource Economists 7: 1113–50. Gilraine, M. 2023. “Air Filters, Pollution, and Student Achievement.” Journal of Human Resources: 54(4): 421–642 Gilraine, M., and A. Zheng. 2022. “Air Pollution and Student Performance in the US.” NBER Working Paper No 30061, National Bureau of Economic Research, Cambridge, MA. Glick, P., and D. E. Sahn. 2010. “Early Academic Performance, Grade Repetition, and School Attainment in Senegal: A Panel Data Analysis.” World Bank Economic Review 24: 93–120. GPE. 2023. “Toward Climate-Smart Education Systems: A 7-Dimension Framework for Action”. Global Partnership for Education Working Paper, Washington DC. Grau, N., D. Hojman, and A. Mizala. 2018. “School Closure and Educational Attainment: Evidence from a Market-Based System.” Economics of Education Review 65: 1–17. Hanushek, E. A., and L. Woessmann. 2020. “The Economic Impacts of Learning Losses”. OECD Education Working Papers No 225, Paris, France. Haverinen-Shaughnessy, U., and R.J. Shaughnessy, 2015. "Effects of classroom ventilation rate and temper- ature on students’ test scores." PloS One 10 (8): e0136165. Heal, C., A. Harvey, S. Brown, A. G. Rowland, and D. Roland. 2022. “The Association between Temperature, Heart Rate, and Respiratory Rate in Children Aged under 16 Years Attending Urgent and Emergency Care Settings.” European Journal of Emergency Medicine 29: 413. 220 The World Bank Research Observer, vol. 39, no. 2 (2024) Hickel, J. 2020. “Quantifying National Responsibility for Climate Breakdown: An Equality-based Attribu- tion Approach for Carbon Dioxide Emissions in Excess of the Planetary Boundary.” Lancet Planetary Health 4: e399–404. Hirabayashi, Y., R. Mahendran, S. Koirala, L. Konoshima, D. Yamazaki, S. Watanabe, H. Kim, and S. Kanae 2013. “Global Flood Risk under Climate Change.” Nature Climate Change 3: 816–21. Holloway, M. S., and E. Rubin. 2022. “Unequal Avoidance: Disparities in Smoke-Induced Out-Migration.” Working Paper, University of Oregon, Eugene, Oregon. Hsiang, S. M., M. Burke, and E. Miguel. 2013. “Quantifying the Influence of Climate on Human Conflict.” Science 341: 1235367. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 IEA. 2020. “Clean Energy Innovation.” https://www.iea.org/reports/energy- technology- perspectives-202 0/clean-energy-innovation. IPCC, 2023. “Summary for Policymakers." In Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPPC, Geneva, Switzerland, pp. 1-34, doi: 10.59327/IPCC/AR6-9789291691647.001 Isaac, F., S. R. Toukhsati, M. Di Benedetto, and G. A. Kennedy. 2021. “A Systematic Review of the Impact of Wildfires on Sleep Disturbances.” International Journal of Environmental Research and Public Health 18: 10152. Jeffries, V., and M. S. Salzer. 2022. “Mental Health Symptoms and Academic Achievement Factors.” Journal of American College Health 70: 2262–5. Joshi, K. 2019. “The Impact of Drought on Human Capital in Rural India.” Environment and Development Economics 24: 413–36. Kaffenberger, M. 2021. “Modelling the Long-Run Learning Impact of the Covid-19 Learning Shock: Actions to (more than) Mitigate Loss.” International Journal of Educational Development 81: 102326. Kemp, L., C. Xu, J. Depledge, K. L. Ebi, G. Gibbins, T. A. Kohler, J. Rockström, M. Scheffer, H.J. Schellnhuber, W. Steffen, and T.M. Lenton 2022. “Climate Endgame: Exploring Catastrophic Climate Change Scenar- ios.” Proceedings of the National Academy of Sciences 119: e2108146119. Köhler-Forsberg, O., H. J. Sørensen, M. Nordentoft, J. J. McGrath, M. E. Benros, and L. Petersen. 2018. “Child- hood Infections and Subsequent School Achievement among 598,553 Danish Children.” Pediatric In- fectious Disease Journal 37: 731–7. Krane, N. K., R. P. DiCarlo, and M. J. Kahn. 2007. “Medical Education in Post-Katrina New Orleans: A Story of Survival and Renewal.” JAMA Network 298: 1052–5. Lagmay, E. A. D., and M. M. T. Rodrigo. 2022. “The Impact of Extreme Weather on Student Online Learning Participation.” Research and Practice in Technology-Enhanced Learning 17: 26. Lai, B. S., A. M. La Greca, C. A. Colgan, W. Herge, S. Chan, J. Medzhitova, M. Short, and B. Auslander 2020. “Sleep Problems and Posttraumatic Stress: Children Exposed to a Natural Disaster.” Journal of Pediatric Psychology 45: 1016–26. Lavy, V., A. Ebenstein, and S. Roth. 2014. “The impact of short term exposure to ambient air pollution on cognitive performance and human capital formation.” NBER Working Paper 20648, National Bureau of Economic Research, Cambridge, MA. Leal Filho, W, S Weissenberger, JM Luetz, J Sierra, I Simon Rampasso, A Sharifi, R. Anholon, J.H.P. Pires, and M. Kovaleva 2023. “Towards a greater engagement of universities in addressing climate change challenges.” Scientific Reports 13 (1): 19030. Lee, T. M., E. M. Markowitz, P. D. Howe, C. - Y. Ko, and A. A. Leiserowitz. 2015. “Predictors of Public Climate Change Awareness and Risk Perception around the World.” Nature Climate Change 5: 1014–20. Li, X., and P.C. Patel, 2021. "Weather and high-stakes exam performance: Evidence from student-level ad- ministrative data in Brazil."Economics letters 199: 109698. Venegas Marin et al. 221 Liu, W., Z. Lian, and Y. Liu. 2008. “Heart Rate Variability at Different Thermal Comfort Levels.” European Journal of Applied Physiology 103: 361–6. May, N. W., C. Dixon, and D. A. Jaffe. 2021. “Impact of Wildfire Smoke Events on Indoor Air Quality and Evaluation of a Low-Cost Filtration Method.” Aerosol and Air Quality Research 21: 210046. McCaul, E. J., G. A. Donaldson, Jr, T. Coladarci, and W. E. Davis. 1992. “Consequences of Dropping out of School: Findings from High School and Beyond.” Journal of Educational Research 85: 198–207. Melo, A. P., and M. Suzuki. 2021. “Temperature, Effort, and Achievement: Evidence from a Large-Scale Stan- dardized Exam in Brazil.” Unpublished manuscript. Meyer, A. 2015. “Does Education Increase Pro-Environmental Behavior? Evidence from Europe.” Ecological Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Economics 116: 108–21. Miller, R. K., and I. Hui. 2022. “Impact of Short School Closures (1–5 days) on Overall Academic Perfor- mance of Schools in California.” Scientific Reports 12: 2079. Miller, S., and M. Vela. 2013. “The Effects of Air Pollution on Educational Outcomes: Evidence from Chile.” IDB Working Paper Series No IDB-WP-468. Inter-American Development Bank, Washington DC. Mora, C., B. Dousset, I. R. Caldwell, F. E. Powell, R. C. Geronimo, C. R. Bielecki, C.W. Counsel, B.S. Dietrich, E.T. Johnston, L.V. Louis, M.P. Lucas, M.M. McKenzie, A.G. Shea, H. Tseng, T.W. Giambelluca, L.R. Leon, E. Hawkins, and C. Trauernicht 2017. “Global Risk of Deadly Heat.” Nature Climate Change 7: 501–6. Mordecai, E. A., S. J. Ryan, J. M. Caldwell, M. M. Shah, and A. D. LaBeaud. 2020. “Climate Change Could Shift Disease Burden from Malaria to Arboviruses in Africa.” Lancet Planetary Health 4: e416–23. Moscoviz, L., and D. K. Evans. 2022. “Learning Loss and Student Dropouts during the covid-19 Pandemic: A Review of the Evidence Two Years after Schools Shut Down.” Center for Global Development Working Paper 609, Washington DC. Mudavanhu, C. 2014. “The Impact of Flood Disasters on Child Education in Muzarabani district, Zim- babwe.” Jàmbá: Journal of Disaster Risk Studies 6(1): 1–8. Mugo, D. 2023. “Nearly Half a Million Children in Malawi Unable to Attend School Due to Cyclone Freddy.” Save the Children. https://www.savethechildren.net/news/nearly-half-million-children-malawi-una ble- attend- school- due- cyclone- freddy. Muttarak, R., and W. Lutz. 2014. “Is Education a Key to Reducing Vulnerability to Natural Disasters and Hence Unavoidable Climate Change? Ecology and Society 19(1): 42 Muttarak, R., and W. Pothisiri. 2013. “The Role of Education on Disaster Preparedness: Case Study of 2012 Indian Ocean Earthquakes on Thailand’s Andaman coast.” Ecology and Society 18(4): 51 Nakitende, A. J., P. Bangirana, N. Nakasujja, J. M. Ssenkusu, C. Bond, R. Idro, Y. Zhao, M. Semrud-Clikeman, and C.C. John 2023. “Severe Malaria and Academic Achievement.” Pediatrics 151: e2022058310. Nordstrom, A., and C. Cotton. 2020. “Impact of a Severe Drought on Education: More Schooling but Less Learning.” Queen’s Economics Department Working Paper, No. 1430, Queen’s University, Department of Economics, Kingston, Ontario. Nübler, L., K. Austrian, J. A. Maluccio, and J. Pinchoff. 2021. “Rainfall Shocks, Cognitive Development and Educational Attainment among Adolescents in a Drought-Prone Region in Kenya.” Environment and Development Economics 26: 466–87. Obradovich, N., R. Migliorini, S. C. Mednick, and J. H. Fowler. 2017. “Nighttime Temperature and Human Sleep Loss in a Changing Climate.” Science Advances 3: e1601555. Odd, D., D. Evans, and A. Emond. 2016. “Preterm Birth, Age at School Entry and Long Term Educational Achievement.” PLoS ONE 11: e0155157. Odera, E. D. 2020. “Thermal Performance of Learning Spaces in Tvet Institutions in Kisumu.” University of Nairobi, Kenya. https://architecture.uonbi.ac.ke/node/633. Padhy, S. K., S. Sarkar, M. Panigrahi, and S. Paul. 2015. “Mental Health Effects of Climate Change.” Indian Journal of Occupational and Environmental Medicine 19: 3. 222 The World Bank Research Observer, vol. 39, no. 2 (2024) Pal, A., T. W. Tsusaka, T. P. L. Nguyen, and M. M. Ahmad. 2023. “Assessment of Vulnerability and Resilience of School Education to Climate-Induced Hazards: A Review.” Development Studies Research 10: 2202826. Palacios, P., and L. Rojas-Velásquez. 2023. “Impact of Weather Shocks on Educational Outcomes in the Municipalities of Colombia.” International Journal of Educational Development 101: 102816. Park, R. J. 2022. “Hot Temperature and High-Stakes Performance.” Journal of Human Resources 57: 400–34. Park, R. J., A. P. Behrer, and J. Goodman. 2021. “Learning Is Inhibited by Heat Exposure, both Internationally and Within the United States.” Nature Human Behaviour 5: 19–27. Park, R. J., J. Goodman, M. Hurwitz, and J. Smith. 2020. “Heat and Learning.” American Economic Journal: Economic Policy 12: 306–39. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Perry, F. B., D. Juan, and L. Dahlin. 2023. “How Are the Children of Pakistan’s 2022 Floods Faring?” World Bank (blog), July 26. https://blogs.worldbank.org/endpovertyinsouthasia/how- are- children- pakistans-202 2- floods- faring. Pichler, A., and E. Striessnig. 2013. “Differential Vulnerability to Hurricanes in Cuba, Haiti, and the Do- minican Republic: The Contribution of Education.” Ecology and Society 18(3): 31 Porras-Salazar, J. A., D. P. Wyon, B. Piderit-Moreno, S. Contreras-Espinoza, and P. Wargocki. 2018. “Reduc- ing Classroom Temperature in a Tropical Climate Improved the Thermal Comfort and the Performance of Elementary School Pupils.” Indoor Air 28: 892–904. Pritchett, L., and A. Beatty. 2015. “Slow Down, You’re Going Too Fast: Matching Curricula to Student Skill Levels.” International Journal of Educational Development 40: 276–88. Psacharopoulos, G., and H. A. Patrinos. 2018. “Returns to Investment in Education: A Decennial Review of the Global Literature.” Education Economics 26: 445–58. Quigley, M. A., G. Poulsen, E. Boyle, D. Wolke, D. Field, Z. Alfirevic, and J.J. Kurinczuk 2012. “Early Term and Late Preterm Birth Are Associated with Poorer School Performance at Age 5 Years: A Cohort Study.” Archives of Disease in Childhood - Fetal and Neonatal Edition 97: F167–73. Randell, H., and C. Gray. 2019. “Climate Change and Educational Attainment in the Global Tropics.” Pro- ceedings of the National Academy of Sciences 116: 8840–5. Reid, C. E., M. Brauer, F. H. Johnston, M. Jerrett, J. R. Balmes, and C. T. Elliott. 2016. “Critical Review of Health Impacts of Wildfire Smoke Exposure.” Environmental Health Perspectives 124: 1334–43. Ridder, N., A. Ukkola, A. Pitman, and S. Perkins-Kirkpatrick. 2022. “Increased Occurrence of High Impact Compound Events under Climate Change.” Npj Climate and Atmospheric Science 5: 3. Rifkin, D. I., M. W. Long, and M. J. Perry. 2018. “Climate Change and Sleep: A Systematic Review of the Literature and Conceptual Framework.” Sleep Medicine Reviews 42: 3–9. Rigaud, K. K., A. De Sherbinin, B. Jones, J. Bergmann, V. Clement, K. Ober, J. Schewe, S. Adamo, B. McCusker, S. Heuser, and A. Midgley 2018. “Groundswell: Preparing for Internal Climate Migration.” Washington DC: World Bank. Ritchie, A., B. Sautner, J. Omege, E. Denga, B. Nwaka, I. Akinjise, S.E. Corbet, S. Moosavi, A. Greenshaw, P. Chue, and X.M. Li 2021. “Long-Term Mental Health Effects of a Devastating Wildfire Are Amplified by Sociodemographic and Clinical Antecedents in College Students.” Disaster Medicine and Public Health Preparedness 15: 707–17. Roach, T., and J. Whitney. 2022. “Heat and Learning in Elementary and Middle School.” Education Eco- nomics 30: 29–46. Sabarwal, S., A. Yi Chang, N. Angrist, and R. D’Souza. 2023. “Learning Losses and Dropouts: The Heavy Cost Covid-19 Imposed on School-Age Children.” In Collapse & Recovery: How COVID-19 Eroded Human Capital and What to Do About It. World Bank, Washington DC. Saif-Ur-Rehman, S. A., and B. Shaukat. 2013. “The Effects of 2010 Flood on Educational Institutions and Children Schooling in Khyber Pukhtoonkhwa: A Study of Charsadda and Swat Districts.” International Journal of Environment, Ecology, Family and Urban Studies 3(3): 1–12. Venegas Marin et al. 223 Santana, O. A., T. P. Silva, O. GSd, S. MMd, I. EdSB, and J. I. Encinas. 2013. “Integration of Face-to-Face and Virtual Classes Improves Test Scores in Biology Undergraduate Courses on Days with Flooding in Brazil.” Acta Scientiarum Education 35: 117–23. Schady, N., A. Holla, S. Sabarwal, and J. Silva. 2023. “Collapse and Recovery: How the Covid-19 Pandemic Eroded Human Capital and What to Do about It.” Washington, DC: World Bank. Schady, N., S. Sabarwal, A. Yi Chang, S. Venegas Marin, R. D’souza, I. Lautharte, I. Tzintzun Valladolid, and L. Schwarz. 2024. “Heat and Learning: How Exposure to Extreme Heat Affects Learning in Brazil.” Washington, DC: World Bank. ( forthcoming). Schmidhuber, J., and F. N. Tubiello. 2007. “Global Food Security under Climate Change.” Proceedings of the Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 National Academy of Sciences 104: 19703–8. Schulze, S. S., E. C. Fischer, S. Hamideh, and H. Mahmoud. 2020. “Wildfire Impacts on Schools and Hospi- tals Following the 2018 California Camp Fire.” Natural Hazards 104: 901–25. Shah, M., and B. M. Steinberg. 2017. “Drought of Opportunities: Contemporaneous and Long-Term Im- pacts of Rainfall Shocks on Human Capital.” Journal of Political Economy 125: 527–61. Sheffield, P. E., S. A. Uijttewaal, J. Stewart, and M. P. Galvez. 2017. “Climate Change and Schools: Environ- mental Hazards and Resiliency.” International Journal of Environmental Research and Public Health 14: 1397. Simmons, S. E., B. K. Saxby, F. P. McGlone, and D. A. Jones. 2008. “The Effect of Passive Heating and Head Cooling on Perception, Cardiovascular Function and Cognitive Performance in the Heat.” European Journal of Applied Physiology 104: 271–80. Singh, A., M. Romero, and K. Muralidharan. 2022. “Covid-19 Learning Loss and Recovery: Panel Data Ev- idence from India.” NBER Working Paper No. 30552, National Bureau of Economic Research, Cam- bridge, MA. Singh, M. K., R. Ooka, H. B. Rijal, S. Kumar, A. Kumar, and S. Mahapatra. 2019. “Progress in Thermal Comfort Studies in Classrooms over Last 50 Years and Way Forward.” Energy and Buildings 188: 149–74. Stanke, C., M. Kerac, C. Prudhomme, J. Medlock, and V. Murray. 2013. “Health Effects of Drought: A Sys- tematic Review of the Evidence.” PLoS Currents 2013: 5. Stenson, C., A. J. Wheeler, A. Carver, D. Donaire-Gonzalez, M. Alvarado-Molina, M. Nieuwenhuijsen, and R. Tham 2021. “The Impact of Traffic-Related Air Pollution on Child and Adolescent Academic Perfor- mance: A Systematic Review.” Environment International 155: 106696. Stott, P. 2016. “How Climate Change Affects Extreme Weather Events.” Science 352: 1517–8. Sun, Q., C. Miao, M. Hanel, A. G. Borthwick, Q. Duan, D. Ji, and H. Li 2019. “Global Heat Stress on Health, Wildfires, and Agricultural Crops under Different Levels of Climate Warming.” Environment Interna- tional 128: 125–36. Taras, H. 2005. “Nutrition and Student Performance at School.” Journal of School Health 75: 199–213. Thamtanajit, K. 2020. “The Impacts of Natural Disaster on Student Achievement: Evidence from Severe Floods in Thailand.” Journal of Developing Areas 54(4): 129–143 Theirworld. 2018. “Safe Schools: The Hidden Crisis.” Thompson, R ., R . Hornigold, L. Page, and T. Waite. 2018. “Associations between High Ambient Tem- peratures and Heat Waves with Mental Health Outcomes: A Systematic Review.” Public Health 161: 171–91. Tian, X., Z. Fang, and W. Liu. 2021. “Decreased Humidity Improves Cognitive Performance at Extreme High Indoor Temperature.” Indoor Air 31:608–27. UNICEF. 2016. “One Week after Hurricane Matthew, at Least 300 Schools Damaged in Haiti, over 100,000 Children Miss Out on Learning.” https://www.unicef.org/press- releases/one- week- after- hurricane- m atthew- least- 300- schools- damaged- haiti- over- 100000. 224 The World Bank Research Observer, vol. 39, no. 2 (2024) ———. 2021a. “One Billion Children at ‘Extremely High Risk’ of the Impacts of the Climate Crisis – UNICEF.” https://www.unicef.org/press- releases/one- billion- children- extremely- high- risk- impac ts- climate- crisis- unicef. ———. 2021b. “Making Climate and Environment Policies for & with Children and Young People.” ———. 2023. “More than 16,000 Children Are Displaced Following Libya Floods – UNICEF.” https://www.unicef.org/press-releases/more- 16000- children- are- displaced%E2%80%AFfollowin g- libya- floods- unicef. Valencia Amaya, M. G. 2020. “Climate Shocks and Human Capital: The Impact of the Natural Disasters of 2010 in Colombia on Student Achievement.” Cuadernos de Economía 39:303–28. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Van der Land, V., and D. Hummel. 2013. “Vulnerability and the Role of Education in Environmentally In- duced Migration in Mali and Senegal.” Ecology and Society 18(4): 14. van Houdt, C. A., J. Oosterlaan, A. G. van Wassenaer-Leemhuis, A. H. van Kaam, and C. S. Aarnoudse-Moens. 2019. “Executive Function Deficits in Children Born Preterm or at Low Birthweight: A Meta-Analysis.” Developmental Medicine & Child Neurology 61: 1015–24. Velásquez, E. E., T. Benmarhnia, J. A. Casey, R. Aguilera, and M. V. Kiang. 2023. “Quantifying Exposure to Wildfire Smoke among Schoolchildren in California, 2006 to 2021.” JAMA Network Open 6: e235863–. Vu, T. M. 2022. “Effects of Heat on Mathematics Test Performance in Vietnam.” Asian Economic Journal 36: 72–94. Wamsler, C., E. Brink, and O. Rentala. 2012. “Climate Change, Adaptation, and Formal Education: The Role of Schooling for Increasing Societies’ Adaptive Capacities in El Salvador and Brazil.” Ecology and Society 17(2): 2 Wang, Q., G. Niu, X. Gan, and Q. Cai. 2022. “Green Returns to Education: Does Education Affect Pro- Environmental Attitudes and Behaviors in China? PloS One 17:e0263383. Wargocki, P., J. A. Porras-Salazar, and S. Contreras-Espinoza. 2019. “The Relationship between Classroom Temperature and Children’s Performance in School.” Building and Environment 157: 197–204. Weems, C. F., L. K. Taylor, N. M. Costa, A. B. Marks, D. M. Romano, S. L. Verrett, and D.M. Brown 2009. “Effect of a School-Based Test Anxiety Intervention in Ethnic Minority Youth Exposed to Hurricane Katrina.” Journal of Applied Developmental Psychology 30: 218–26. Wen, J., and M. Burke. 2022. “Lower Test Scores from Wildfire Smoke Exposure.” Nature Sustainability 5: 947–55. Williams, P. C., A. W. Bartlett, A. Howard-Jones, B. McMullan, A. Khatami, P. N. Britton, and B.J. Marais 2021. “Impact of Climate Change and Biodiversity Collapse on the Global Emergence and Spread of Infectious Diseases.” Journal of Paediatrics and Child Health 57: 1811–8. Williamson, K., A. Satre-Meloy, K. Velasco, and K. Green. 2018. Climate Change Needs Behavior Change: Making the Case for Behavioral Solutions to Reduce Global Warming. Arlington, VA, USA: Rare. WMO. 2021. “Weather-Related Disasters Increase over Past 50 Years, Causing More Damage but Fewer Deaths.” https://wmo.int/media/news/weather-related-disasters-increase-over-past-50-years-causi ng- more- damage- fewer- deaths. World Bank. 2022a. “Guide for Learning Recovery and Acceleration.” https://thedocs.worldbank.org/en /doc/523b6ac03f2c643f93b9c043d48eddc1-0200022022/related/English-Exec-Summary-Guide-for- Learning-Recovery- and- Acceleration- Final.pdf. ———. 2022b. “Rainwater Harvesting in Yemen: A Durable Solution for Water Scarcity.” ———. 2024. “Global Program for Safer Schools (GPSS).” Wu, G. 2022. “Do Wildfires Harm Student Learning?” Haslam College of Business Working Paper #2022-02, University of Tennessee, Department of Economics, Knoxville, Tennesse. Venegas Marin et al. 225 Yang, W., and H. J. Moon. 2019. “Combined Effects of Acoustic, Thermal, and Illumination Conditions on the Comfort of Discrete Senses and Overall Indoor Environment.” Building and Environment 148: 623–33. Yang, Y., L. Hu, R. Zhang, X. Zhu, and M. Wang. 2021. “Investigation of Students’ Short-Term Memory Per- formance and Thermal Sensation with Heart Rate Variability under Different Environments in Sum- mer.” Building and Environment 195: 107765. Yeganeh, A. J., G. Reichard, A. P. McCoy, T. Bulbul, and F. Jazizadeh. 2018. “Correlation of Ambient Air Tem- perature and Cognitive Performance: A Systematic Review and Meta-Analysis.” Building and Environ- ment 143: 701–16. Downloaded from https://academic.oup.com/wbro/article/39/2/177/7642013 by World Bank user on 19 July 2024 Zhang, X., X. Chen, and X. Zhang. 2018. “The Impact of Exposure to Air Pollution on Cognitive Perfor- mance.” Proceedings of the National Academy of Sciences 115: 9193–7. ———. 2024.“Temperature and Low-Stakes Cognitive Performance.” Journal of the Association of Environ- mental and Resource Economists 11: 75–96. Zhu, H., H. Wang, Z. Liu, D. Li, G. Kou, and C. Li. 2018. “Experimental Study on the Human Thermal Comfort Based on the Heart Rate Variability (hrv) Analysis under Different Environments.” Science of the Total Environment 616: 1124–33. Zivin, J. G., and J. Shrader. 2016. “Temperature Extremes, Health, and Human Capital.” Future of Children, 26(1): 31–50. Zivin, J. G., Y. Song, Q. Tang, and P. Zhang. 2020. “Temperature and High-Stakes Cognitive Performance: Ev- idence from the National College Entrance Examination in China.” Journal of Environmental Economics and Management 104: 102365. Zuilkowski, S. S., M. C. Jukes, and M. M. Dubeck. 2016. ““I Failed, No Matter How Hard I Tried”: A Mixed- Methods Study of the Role of Achievement in Primary School Dropout in Rural Kenya.” International Journal of Educational Development 50: 100–7. The World Bank Research Observer © 2024 International Bank for Reconstruction and Development / The World Bank. Published by Oxford University Press https://doi.org/10.1093/wbro/lkae001 226 The World Bank Research Observer, vol. 39, no. 2 (2024)