Instrument breadth, function span and design depth in coal-power transitions: processed data, code and reproducibility materials

This record provides the processed data, executable code, locked numerical outputs and scientific figure source files accompanying the manuscript “Instrument breadth, function span and design depth in coal-power transitions”. The study distinguishes instrument breadth, function span and conditional design depth in electricity-policy portfolios and relates these constructs to later coal-electricity shares across 49 economies. The archive supports standalone processed-input-to-result reproduction. It contains a processed policy panel exposing all six primary instrument-score fields and function activations; a separate coal-model panel; five executable analysis and figure scripts; a one-command run-all workflow; exact tested software requirements; 2,000-draw country-block decomposition outputs; 4,999-draw restricted wild-cluster bootstrap-t inference; leave-one-country diagnostics; function-span construct validation; figure-source data; locked numerical outputs; and SHA-256 integrity manifests. The exact identity aggregate stringency = instrument breadth × conditional design depth is reproduced from the supplied processed inputs. Function span is treated as a separate construct-differentiation measure and is not included in the identity. Its coal-share estimates are statistically uncertain and approach zero after wider-policy adjustment. All reported associations are observational and are not interpreted as causal effects. The package requires no internet connection, Google Drive, Colab upload, previous archive or author-specific path. Running “python run_all.py” automatically compares every locked CSV byte-for-byte with its reproduced counterpart and checks locked PNG/TIFF figures pixel-for-pixel. Upstream acquisition and harmonization of public OECD CAPMF, Ember and World Development Indicators source data are outside the claimed reproduction scope.