Conference Agenda

Formal education plays a crucial role in human capital accumulation and economic growth. Previous research shows that heat inhibits learning, with negative impacts on learning outcomes observed on days hotter than 26.7⁰C. However, these findings often do not account for acclimatization, adaptive behaviors, or potential non-linear relationships between temperature and learning outcomes. This study addresses these gaps. To this end, we link test scores and household data from 370,585 students across more than 120 regions worldwide who participated in the OECD’s PISA (Program for International Student Assessment) study with regional temperature extremes from the ERA5 climate reanalysis dataset. We focus on a relative heat stress index, classifying days as hot when their maximum temperature exceeds the regional 95th percentile threshold (TX95p) during the academic years preceding the PISA test. Through this approach, we examine the long-term effects of heat exposure on learning outcomes. Our findings indicate that hot days have a significant negative impact on test scores, with the marginal effect of the first hot day on average scores being -0.025 standard deviations (SD) of scores with diminishing impact of additional hot days. We also find that compared to the Average scores, the Reading scores are the most negatively affected (-0.032 SD, 28% larger), followed by Mathematics (-0.023 SD, 8% smaller) and Science (-0.015 SD 40% smaller). We further provide the first evidence that the marginal effect of hot days is heterogeneous across the student performance spectrum, with lower-scoring students being more negatively affected than their higher-performing counterparts. Our results confirm that increasing climate extremes will impair formal learning, highlighting the greater vulnerability of weaker students. Incorporating these findings into economic models is essential for understanding how climate extremes may hinder human capital formation and, consequently, long-term economic growth.