EXTREME WEATHER EVENTS AND ECONOMIC VULNERABILITY: MICRO-EVIDENCE FROM FIRMS AND HOUSEHOLDS

Abstract Chapter 2: "The Impact of Temperature and Heat Waves on Italian Manufacturing Firms" - The study aims to assess how high temperatures and heat waves affect productivity, output, and input usage among Italian manufacturing firms. Using firms' financial accounts, we estimate non-linear effects of temperature on firm-level output, total factor productivity, and input supply. We show that days with a temperature above 32°C and heat waves significantly decrease firms' productivity and lead to a corresponding output decline in both the short and long-run, predicting extensive future economic damages. Similarly, heat waves are also found to significantly impact output and productivity, with substantial long-lasting effects. Analysis of future scenarios shows that predicted temperature impacts are likely to be linked to major output and productivity losses under pessimistic climate change scenarios, but also under more optimistic conditions, albeit with considerable local heterogeneity, with industries in the south and flat areas of the north being the most affected. Examining how firms adapt to these challenges and identifying sectoral vulnerabilities, we highlight the need for climate adaptation policies to strengthen resilience in Italy's manufacturing sector amid increasing climate risk. Abstract Chapter 3: "Household perceptions and responses to climate shocks, evidence from Tanzania" - This paper investigates how households perceive and recall climate shocks, and whether recall accuracy conditions their resilience to extreme weather events. We combine three waves of the Tanzania National Panel Survey (2008-2013) with agro-meteorological measures of droughts and floods to construct an objective benchmark of shock exposure. We then analyse the behavioral determinants of recall accuracy, focusing on recency bias, experiential learning, salience effects from contemporaneous weather conditions, and the moderating role of education and livelihood strategies. Our findings reveal systematic biases in recall. Households are less likely to identify shocks as time passes, yet repeated exposure improves accuracy. Education mitigates these biases, particularly for non-farmers, and farmers recall shocks more accurately than non-farmers, consistent with their greater dependence on weather variability. In the second part of the analysis, we show that recall accuracy has substantive welfare implications. While climate shocks generally depress consumption growth, losses are concentrated among households that misperceive shocks, especially systematic under-reporters, with under-reporting farmers emerging as the most vulnerable group. By contrast, households with accurate recall show resilience, as neither contemporaneous nor lagged shocks significantly reduce their consumption growth. These results demonstrate that behavioral frictions in memory and perception amplify the economic costs of climate shocks, while accurate recall dampens them. The findings highlight recall accuracy as an economically meaningful household characteristic and underscore the importance of integrating behavioral insights into the analysis of climate vulnerability and resilience.