Does Multilevel Governance Matter for Poverty Reduction? A Sequential Explanatory Mixed-Methods Analysis of SDG 1 Implementation in Subnational Indonesia

Abstract

Persistent poverty in metropolitan peripheries reveals a proximity without prosperity paradox — regions that are geographically integrated into urban economic cores yet structurally excluded from their developmental benefits. This study examines whether Multilevel Governance (MLG) configuration influences the effectiveness of SDG 1 (No Poverty) implementation through serial mediation by inter-governmental coordination and collaborative governance in West Bandung Regency (Kabupaten Bandung Barat), Indonesia. Despite its proximity to the Bandung Metropolitan Area — which contributes approximately 15% of West Java's gross domestic product — West Bandung Regency records a persistent poverty rate of 9.87%, substantially above the provincial average. Employing a sequential explanatory mixed-methods design, the study integrates Partial Least Squares Structural Equation Modelling (PLS-SEM) with bootstrapping (5,000 resamples, n = 400) and qualitative NVivo analysis of 40–45 key informants. Nine hypotheses testing direct, indirect, and serial mediation pathways are evaluated. Findings confirm that MLG exerts a significant positive direct effect on SDG 1 outcomes, substantially mediated through inter-governmental coordination quality and collaborative governance, with the full serial path MLG → Coordination → Collaborative Governance → SDG 1 proving statistically significant. Principled engagement emerges as the most dominant collaborative dimension. These results advance MLG theory by empirically specifying the institutional transmission mechanisms through which governance architecture translates into poverty reduction outcomes in asymmetric decentralization contexts.