Water and sanitation investment efficiency in Latin America and the Caribbean: A dynamic SBM-DEA analysis (2012–2021) — data and code

Data and Python code to reproduce the results of the article "Water and sanitation investment efficiency in Latin America and the Caribbean: A dynamic SBM-DEA analysis (2012–2021)". The study estimates a non-oriented dynamic slacks-based measure (DSBM) model (Tone and Tsutsui, 2010) with variable returns to scale for 11 Latin American and Caribbean countries (Argentina, Brazil, Chile, Colombia, Costa Rica, Dominican Republic, Guyana, Mexico, Paraguay, Peru, Trinidad and Tobago) over 2012–2021. Inputs: water and sanitation investment (% of GDP, INFRALATAM) and water-use intensity (freshwater withdrawals per US$1,000 of GDP at constant 2015 prices, World Bank). Desirable outputs: safely managed drinking water and sanitation (SDG 6.1.1 and 6.2.1, WHO/UNICEF JMP). Undesirable output: water stress (SDG 6.4.2, World Bank). Desirable carry-over: water and sanitation infrastructure stock (% of GDP), constructed by perpetual inventory at constant prices (δ = 0.04). The package also reports dynamic super-efficiency scores (Tone, 2002) and twelve alternative specifications. Contents: raw input data (data/raw), analytical panel and imputation records (data/processed), Python scripts (code), and result tables and figures (results). The file Zenodo_Art3_full_package.zip contains all files in the folder structure expected by the scripts; see README.md for instructions and for the correspondence between result files and the tables and figures of the article. Code is released under the MIT licence; derived data and results under CC BY 4.0. The original series remain subject to the terms of INFRALATAM (IDB, CAF and ECLAC), the World Bank World Development Indicators and the WHO/UNICEF Joint Monitoring Programme, which must be cited together with this dataset.