Abstract
This paper estimates the short-run economic impacts of low-intensity typhoons—tropical storms and depressions—on local activity in Vietnam. Leveraging monthly electricity consumption data for over two million households across three coastal provinces from 2012–2022, we construct a high-resolution panel merged with typhoon track data. We exploit hyperlocal variation in typhoon exposure using a dynamic event study and a stacked difference-in-differences design, defining treatment based on proximity within 17 kilometers of a storm’s path. Results show a sharp and statistically significant decline in electricity consumption—up to 5% in the month of landfall—concentrated in areas directly hit by tropical storms. Tropical depressions exhibit smaller, noisier effects. Comparison with BRDF-corrected nighttime lights (NTL) data reveals that electricity consumption provides more reliable estimates of typhoon impacts at lower administrative levels, as NTL measures fail the parallel trends test. Our findings underscore the economic relevance of frequently overlooked, lower-intensity storms and highlight the value of granular energy data in assessing disaster-induced disruptions in developing country contexts.