Replication Materials for “When Extreme Rainfall Dims Local Economies"
Description
This repository contains replication materials for the manuscript “When Extreme Rainfall Dims Local Economies: Nighttime-Light Reductions, Food-Insecurity Outcomes, and Household Borrowing in Bangladesh.” The study examines how extreme rainfall is associated with short-run local economic activity and complementary household welfare outcomes in Bangladesh. The local economic analysis uses a balanced district-month panel covering all 64 districts over 132 months from 2015 to 2025. Monthly nighttime-light intensity from VIIRS is used as a proxy for local economic activity and is combined with CHIRPS precipitation data to construct extreme-rainfall exposure measures and dynamic rainfall histories. The household analysis uses the Bangladesh Household Income and Expenditure Survey (HIES) 2022 to examine associations between recent district-level extreme-rainfall exposure and food-insecurity and household-borrowing outcomes. The empirical framework includes district and time fixed effects, distributed rainfall timing specifications, household-level regression models, alternative exposure windows, placebo and robustness analyses, multiple-testing adjustments, and supplementary inference procedures. The public replication archive contains Stata 14-compatible analysis code, the district-month climate and nighttime-light panel, district-level exposure files used for the household analysis, statistical tables, figures, diagnostics, documentation, and result-verification materials. Original household-level HIES 2022 microdata and derived household-level analytical files are not redistributed and must be obtained through the authorized Bangladesh Bureau of Statistics data-access process. The package therefore reproduces the district-month analysis directly from the deposited data and reproduces the household analysis when authorized HIES 2022 source files are supplied in the documented input location. Third-party source materials remain subject to the terms of their respective data providers.
Files
Steps to reproduce
Download and extract the replication ZIP while preserving the folder structure and read the separately deposited README.txt before running the analysis. The replication code requires Stata 14.0 or later. Set the Stata working directory to the extracted package root and run do "02_stata/00_RUN_FULL_REPRODUCTION_STATA14.do". The master script reproduces the district-month analysis using the deposited climate and nighttime-light data. The household component uses Bangladesh HIES 2022 microdata, which are not redistributed with this repository. Researchers wishing to reproduce the household analysis must obtain authorized HIES 2022 files from the Bangladesh Bureau of Statistics and place the required source modules in the input location specified in 08_hies_integration/01_raw_hies/README_RESTRICTED_HIES_INPUT.txt. When the authorized files are present, the master workflow constructs the household analytical data, merges district-level rainfall exposure histories, and reproduces the food-insecurity and borrowing analyses, robustness checks, and associated outputs. If the HIES source files are absent, the district-month replication can still be completed independently.
Institutions
- Jagannath UniversityDhaka Division, Dhaka