Replication Package for: "How Much Education Spending Is Enough? Threshold Effects of Public Education Expenditure on Economic Growth in Sub-Saharan Africa"

This repository contains the complete replication package for the paper "How Much Education Spending Is Enough? Threshold Effects of Public Education Expenditure on Economic Growth in Sub-Saharan Africa" by Ernest Pele Botsyoe. The paper examines whether the growth effect of public education expenditure in Sub-Saharan Africa is regime-dependent rather than linear. Using Hansen's (1999) fixed-effects panel threshold regression (Stata: xthreg) across 47 SSA countries over 1970–2023 — the first application of this framework to education expenditure in SSA at this scale — the study identifies a statistically significant spending threshold of 3.14% of GDP, below which education expenditure yields no reliably estimable growth dividend, and above which positive growth returns activate sharply. The threshold is not gradual: it constitutes a structural regime boundary. Colonial heritage shifts the threshold across linguistic groups: Francophone SSA economies cross into the high-return regime at 4.91% of GDP, while Anglophone SSA requires 6.40% — a divergence attributed to CFA franc zone monetary architecture and colonial institutional inheritance. Results are confirmed under alternative specifications and System GMM instrumentation. The package includes: Panel dataset (47 SSA countries, 1970–2023) with all variables as defined in the paper Stata do-files for Hansen (1999) single- and double-threshold estimation (xthreg) System GMM robustness scripts with instrument validity diagnostics Subgroup estimation files for Francophone and Anglophone SSA Output tables corresponding to all results reported in the manuscript Data sources: World Bank World Development Indicators (WDI); World Bank Worldwide Governance Indicators (WGI). Keywords: public education expenditure; economic growth; panel threshold regression; Hansen (1999); Sub-Saharan Africa; colonial legacy; CFA franc zone; Francophone; Anglophone; nonlinearity JEL Classification: O40; H52; C23; I25; O55