Rising Temperature and International Trade Dynamics

This repository contains the replication package for the paper “Rising Temperature and International Trade Dynamics”. It provides analysis-ready Stata datasets and do-files that reproduce the main empirical results, robustness checks, channel tests, heterogeneity analyses, and extended analyses reported in the paper (Tables 1–15 and Figures 1–3).

Contents

Datasets (Stata .dta):

Panel A main.dta – core city-by-month panel with trade growth outcomes and local climate variables. Key variables include export/import growth (dlnex, dlnim), price and output changes (dlncpi, dlnind), and climate measures (tem, pre, sun, hum) with constructed lags (ltem, l2tem, etc.).

Panel B heterogeneity.dta – extended outcomes and subsamples for heterogeneity tests, including domestic vs. foreign trade indicators (dlnex1–dlnex5, dlnex2, dlnim2) and product-level outcomes used to build Figure 3.

Panel C extended.dta – variables for export structure and diversification analyses, including pm, nep, lc, pdiv, sdiv and external temperature shock measures (extem and lags).

Code (Stata do-files):
01_baseline_regression.do, 02_robustness_checks.do, 03_channel_analysis.do, 04_heterogeneity_analysis.do, 05_extended_analysis.do.

Replication guide: README.pdf (software requirements, variable notes, and execution steps).

Software requirements
Stata ≥15 (tested on Stata 16/17). Required packages: reghdfe and estout (install via ssc install reghdfe and ssc install estout).

How to run
Set the working directory to the folder containing the data and do-files, then run do-files in order:

01_baseline_regression.do

02_robustness_checks.do

03_channel_analysis.do

04_heterogeneity_analysis.do

05_extended_analysis.do

These files generate the regression outputs used in the manuscript. Figures 1–2 are plotted from reported bin estimates; Figure 3 is constructed from cumulative effect estimates across products (see README for details).