Government expenditure and shadow economy in Africa

Purpose The standard Keynesian theory predicts that government spending can stimulate the official economy through aggregate demand. However, the relationship seems to be nonlinear, as predicted empirically in accordance with the Armey curve hypothesis. Furthermore, recent events such as asymmetric information, transaction costs, regime shifts, and data-generating processes, among others, could cause government expenditure to exhibit nonlinear and asymmetric behavior. Thus, this study extends the literature to investigate the nonlinear nexus between government spending and the shadow economy in Africa. Design/methodology/approach The study draw data from 30 African countries spanning 1996–2020. The two-step system generalized method of moments (2SGMM), fully modified ordinary least squares augmented mean group and dynamic common correlated effect mean group (DCCEMG) are used as the estimation techniques. Findings We demonstrate an inverted U-shaped relationship between the shadow economy and government spending using advanced panel econometric techniques that account for the econometric pitfalls of reverse causality, endogeneity, heterogeneity, and cross-sectional dependence. The study further obtained a threshold value of 36.146% of GDP for government expenditure. Research limitations/implications The implication of the study suggests that government spending may serve as a fiscal tool for downgrading the extent of shadow economy after a certain threshold. Originality/value This study provides fresh insights into the nexus between government expenditure and shadow economy, and also tested the Armey hypothesis in the linkages.