Fiscal Capacity and Climate Finance Effectiveness in Sub-Saharan Africa

Unofficial AI-generated summary based on the public title and abstract. Not an official translation.

🔬研究者:A novel dataset and robust empirical strategy to study climate finance effectiveness, useful for replication and extension to other regions.

🏢実務担当者:Development banks and climate funds can use these findings to target fiscal capacity building alongside financial disbursements.

🏛政策担当者:Donor countries and recipient governments should prioritize strengthening fiscal governance to maximize climate finance outcomes.

Supplementary Materials Description Fiscal Capacity and Climate Finance Effectiveness in Sub-Saharan Africa 1. Dataset The supplementary data file (climate_finance_v3_data_2002_2023.xlsx) contains panel data for 40 Sub-Saharan African countries covering the period 2002 to 2023. The dataset includes 880 country-year observations across 33 variables, structured for replication of all analyses reported in the manuscript. 1.1 Variable Categories The variables are organised into the following categories: Category Variables Source Outcome variables SDG Index, HDI, Resilience Index, Energy Transition Index, Adaptive Capacity Index Sachs et al. (2024); UNDP (2024); author-constructed Climate finance Total climate finance, adaptation finance, mitigation finance (all as % of GDP) OECD CRS; Climate Funds Update Fiscal-governance capacity Fiscal-Governance Capacity Index (composite of tax revenue, government effectiveness, fiscal space, public investment intensity) IMF GFS; World Bank WGI; WDI Controls GDP per capita, population, trade openness, urbanisation, inflation, debt-to-GDP World Bank WDI; IMF WEO Interaction terms Climate finance x fiscal-governance capacity, adaptation x capacity, mitigation x capacity Author-constructed Shock variables Climate shock exposure, climate vulnerability, resource dependence, fragile state indicator Author-constructed; World Bank 1.2 Key Features of the Dataset •   The Resilience Index (Version 3) excludes the fiscal-governance capacity index to avoid circularity. It is constructed from electricity access (35%), renewable energy (25%), life expectancy (25%), and gross fixed capital formation (15%), normalised to a 0 to 100 scale. •   The Fiscal-Governance Capacity Index is a composite of four standardised components (tax revenue/GDP, government effectiveness, fiscal space, GFCF/GDP), averaged with equal weights and normalised to 0 to 100. •   Climate finance is measured as commitments (not disbursements) expressed as a percentage of recipient GDP, harmonised from OECD CRS and Climate Funds Update data. •   All monetary variables are in constant 2015 USD. Inflation is log-transformed as ln(1 + inflation). •   Missing values in GFCF and public investment are imputed using country-means followed by global means where necessary. 1.3 Coverage Countries: 40 Sub-Saharan African nations including all major economies (Nigeria, South Africa, Kenya, Ghana, Ethiopia) and smaller states (Lesotho, Eswatini, Eritrea). Year range: 2002 to 2023. Panel is unbalanced due to data availability constraints in conflict-affected and small states. 2. Replication Codebook The replication codebook (replication_codebook_v3.pdf) provides complete Python code to reproduce all analyses in the manuscript. The codebook is organised into sequential sections corresponding to the tables and figures in the paper. 2.1 Structure of the Codebook Section Content Output 1. Setup Data loading, library imports Loaded dataset 2. Index construction Fiscal-governance capacity index construction (z-scores, equal weights, normalisation) Table 1 descriptive statistics 3. Outcome variables Resilience index, energy transition index, climate shock exposure Alternative dependent variables 4. Interactions and lags Core interaction terms, temporal lags, spatial lag instrument Variables for all models 5. Main analysis Pooled OLS, entity fixed effects, two-way fixed effects, IV/2SLS Tables 2 and 3 6. Mechanisms Public investment, electricity access, human capital as intermediate outcomes Table 5 (first three columns) 7. Alternative outcomes Resilience index and energy transition index regressions Table 5 (last two columns) 8. Robustness Winsorised, subsample (post-2010, post-2015, excluding South Africa), non-fragile Table 7 2.2 Software Requirements The replication code requires Python 3.8 or later with the following packages: pandas, numpy, statsmodels, linearmodels, and scipy. All packages are open-source and freely available via pip or conda. 2.3 How to Replicate •   Place the data file (climate_finance_v3_data_2002_2023.xlsx) in the working directory. •   Run the code sections sequentially; each section builds on variables created in prior sections. •   Key results are printed to the console with coefficient estimates, standard errors, and significance levels. •   The expected results summary table at the end of the codebook provides benchmark values for verification. 2.4 Notes on the IV Estimation The IV/2SLS specification instruments climate finance with its first temporal lag and a spatial lag (regional average climate finance excluding the country itself). The first-stage F-statistic of 370.03 exceeds the Stock-Yogo critical value for 10% maximal IV size. The code implements manual 2SLS to avoid collinearity issues that can arise with packaged IV routines when instrumenting interaction terms.

• Zenodo https://zenodo.org/records/20408388first seen 2026-05-28 04:15:45 · last seen 2026-05-30 04:26:45