Climate Policy Diffusion

This repository contains the code and data used for the analysis in the paper "The international diffusion of climate policy: Theory and evidence". The paper is available here.

This paper investigates mechanisms of climate policy diffusion across national jurisdictions in the context of the Paris Agreement.

Abstract

Globally coordinated climate action has resulted in suboptimal GHG emission reductions and unilateral, second-best, climate policies have so far provided the bulk of these reductions. Using an open economy general equilibrium framework, we propose that the adoption of climate policy is partly determined by a process of policy diffusion whereby actions of foreign jurisdictions affect domestic conditions and policy decisions. We focus on diffusion mechanisms related to (i) access to improved foreign abatement technology and (ii) policy adoption by foreign jurisdictions. We apply our framework to the adoption of feed-in tariffs (FiT), renewable portfolio standards (RPS) and carbon pricing mechanisms. Overall, results highlight differences among policies. The evidence suggests that improved access to climate change mitigation technologies leads to earlier adoption of RPS and carbon taxes but not FiT or an emissions trading system (ETS). It also suggests that countries with common legacy institutions influence each other's adoption decisions in the case of FiT.

Repository layout

• _code/: analysis and pipeline scripts (Python-first)
• data/raw/: raw inputs (ignored by git; see data/README.md)
• data/processed/: intermediate outputs (ignored by git)
• _dataset/: finalized analysis datasets used by models
• output/: figures/tables
• config/: path and data source configuration
• doc/: documentation (see doc/README.md)

Data refresh workflow

1. Place raw files in data/raw using the paths listed in data/README.md.
2. (Optional) Download World Bank WDI bulk data via:

PYTHONPATH=_code python3 -m pipeline.ingest_wdi

3. Use config/paths.json to override default directories if needed.
4. Run a preflight check for raw inputs:

PYTHONPATH=_code python3 -m pipeline.build_dataset

5. Build datasets:

PYTHONPATH=_code python3 -m pipeline.data_processing

To run a partial stage, use:

PYTHONPATH=_code python3 -m pipeline.data_processing --stage controls

Documentation (mkdocs)

To preview the docs locally:

mkdocs serve

Notes on data sources

Some sources are downloadable via API (e.g., World Bank WDI) while others are manual due to licensing or authentication (IEA, CEPII, V-Dem, etc.). The canonical list lives in config/sources.json.