Nonlinear Effects of Digital Governance on Sustainable Development: Panel Threshold Analysis for EU Member States

Abstract

This study aims to determine whether and how institutional quality conditions the effect of e-government adoption on sustainable development in the EU. This paper examines the nonlinear relationship between digital governance and sustainable development performance in the European Union. Using a balanced panel of 27 EU member states over 2015-2023 $(\mathrm{N}=243)$, we apply Hansen's panel threshold regression to test whether institutional quality conditions the effectiveness of e-government adoption in achieving the Sustainable Development Goals (SDGs). The results reveal a statistically significant “Digital Governance Paradox”: the effect of e-government usage on SDG outcomes varies by institutional quality regime. For SDG 9 (Innovation and Infrastructure), e-government has a strong positive effect in countries with low government effectiveness $(\beta=+0.104, \mathrm{p}< 0.01)$ but a negative effect in high-quality-institution countries $(\beta=-0.065, \mathrm{p}<0.01)$. For SDG 16 (Institutions), the pattern reverses: e-government undermines institutional outcomes in weak-institution settings $(\beta=-0.056, \mathrm{p}<0.01)$ but strengthens them in strong-institution countries $(\beta=+0.038, \mathrm{p}<0.01)$. The Chow test confirms structural heterogeneity between EU15 and EU13 cohorts $(\mathrm{p}<0.001)$. These findings carry direct implications for the EU Digital Decade Policy Programme 2030.