Education and the Shadow Economy: A Panel Cointegration Analysis

Abstract

This study examines the impact of education on the shadow economy across 132 countries over the period 1991–2020. To this end, panel cointegration and panel error-correction models are employed. The results of the panel cointegration tests indicate a stable long-term relationship among the selected variables. Both short- and long-run coefficients are estimated using a fully modified ordinary least squares (FMOLS) panel regression. The findings reveal a positive association between education and the size of the shadow economy. To account for institutional factors and potential cross-country heterogeneity, we construct subsamples based on countries' levels of corruption and economic freedom. Finally, the study applies a novel panel Granger noncausality approach (JKS test) to explore the bidirectional relationship between education and the shadow economy. The results indicate a causal relationship only in countries with high levels of corruption. Overall, the empirical evidence provides valuable insights for researchers and policymakers, suggesting that policies aimed at expanding higher education may have unintended adverse effects on economic growth by contributing to an increase in the shadow economy.