Public Education Expenditure and Economic Growth in Latin America: A Panel Data Analysis of Contemporaneous and Lagged Effects, 2000–2022

Abstract

This study examines the relationship between public expenditure on education and economic growth across sixteen Latin American countries over the period 2000–2022, using a balanced-to-unbalanced macro panel drawn entirely from the World Bank's World Development Indicators. Motivated by the long-standing "Latin American growth puzzle", in which the region's early advantage in schooling failed to translate into sustained economic convergence, the analysis estimates pooled, fixed-effects, and random-effects specifications of an augmented growth model, and probes the temporal structure of the education–growth link through distributed-lag estimation. A Hausman test marginally favours the random-effects estimator, under which education expenditure is positively and significantly associated with the log of GDP per capita, with the association strongest at a lag of one to two years (coefficients of roughly 0.06–0.065, significant at the 1% level). Under the more demanding two-way fixed-effects estimator, however, the coefficient shrinks toward zero and loses conventional significance, becoming only marginally significant once Driscoll–Kraay standard errors are used to accommodate the pronounced cross-sectional dependence detected in the residuals. The evidence therefore points to a fragile and estimator-sensitive positive relationship rather than a robust growth dividend from educational spending alone. These findings are consistent with a body of Latin American research that emphasises the effectiveness, rather than the mere volume, of educational investment (Arias Ortiz et al., 2023; Hanushek, 2013), and they caution against treating higher education budgets as a sufficient instrument for growth in the absence of complementary improvements in the quality and institutional efficiency of spending.