Estimating medium-term regional monthly economic activity reductions during the COVID-19 pandemic using nighttime light data

Abstract

Economic impact estimates of the initial lockdowns due to the COVID-19 pandemic showed a significant reduction in economic activities globally. However, the succeeding impacts and their spatiotemporal distribution within countries remain unknown. Studies showed that nighttime light data (NTL) has effectively revealed the spatiotemporal dimensions of the economic effects of COVID-19. Thus, this study used NTL data to determine the medium-term regional monthly economic impacts of the pandemic in the Philippines in terms of the Economic Activity Reduction (EAR) index. We generated a spatial error model, regressing pre-pandemic NTL on mean temperature, maximum rainfall, and built-up area. This model explained 81.6% of the pre-pandemic NTL and was used to estimate the counterfactual NTL. We subtracted the actual from the counterfactual to compute the EAR. Then, the EAR was regressed on regional factors to determine which ones influence the impacts. Results showed uneven distribution of EAR across space and time. The EAR was generally higher in urban regions than in rural ones. Overall, more regions in the south had higher EAR. Temporally, the EAR showed a dynamic pattern, increasing in less urban regions and decreasing in highly urbanized regions. Regional analysis showed that urbanization level, population density, and poverty incidence had a significant positive relationship with the EAR. Beyond the immediate impacts, NTL effectively revealed spatiotemporal dimensions of the economic effects of a long-term global hazard.