Replication materials for: Beyond tariff adjustment: Fuel-price shocks and utility financial recovery in Korea
Description
Korean electricity prices, sales and utility financial records for 2011-2024, with Python scripts that reproduce every table and figure in the related article. The data cover Korea Electric Power Corporation (KEPCO), Korea Hydro & Nuclear Power (KHNP) and the five thermal generation subsidiaries (KOEN, KOMIPO, KOSPO, KOWEPO and EWP). Values are taken from official publications and databases: KEPCO's Statistics of Electric Power in Korea; annual business reports, including amended reports, filed with the Financial Supervisory Service through the DART electronic disclosure system; the Korea Power Exchange (KPX) Electric Power Statistics Information System (EPSIS) and KPX's official annual system marginal price; Korea Gas Corporation (KOGAS) gas tariffs for power generation; the Korea Energy Economics Institute (KEEI) Yearbook of Energy Statistics; the Bank of Korea Economic Statistics System (ECOS); and Statistics Korea. Each of the 1,380 ledger values carries a provenance record in data/source_catalog.csv giving the publisher, document, direct link, and the table, account or query it came from. The smaller input files carry their own source columns. The archive contains extracted values and the authors' analysis code. The original corporate reports and statistical yearbooks are not redistributed; the source catalogue links to their publishers. FILE_GUIDE.md lists every file in the archive and explains how to run the analysis.
Files
Steps to reproduce
Python 3.12. Extract the archive and open the folder up_financial_recovery. Then run: python -m pip install -r requirements.txt python reproduce.py Results are written to results/: 13 table files matching Tables 1-5 and Supplementary Tables S1-S8, three figures in PNG, PDF and SVG, the plotted values behind each figure, and the intermediate series. To run with the Python standard library only, without plotting: python reproduce.py --no-figures --output-dir results_numbers reproduce.py checks every input file against the SHA-256 digests in input_manifest.json before calculating, and stops if an input has changed. The figures use Times New Roman; where that font is not installed, matplotlib substitutes its default font and prints "findfont" warnings, which do not affect any number.
Categories
Funders
- Kepco International Nuclear Graduate School