Urban tourism in underdeveloped areas: city growth generates insignificant spillover effects

Urban tourism (UT) is increasingly expected to stimulate regional growth in resource-rich yet economically underdeveloped areas, but how regional disparities evolve and what drives them remain insufficiently understood. This study investigates UT development across Guangxi, China (2010–2022) through a causally linked evaluation system that integrates the DPSIR model with economic efficiency theory. We further combine entropy-weighted TOPSIS, ensemble learning algorithms, and GTWR to simultaneously quantify factor importance hierarchies and their spatiotemporal non-stationarity through the triple bottom line lens. The findings reveal a persistent core–periphery pattern in which Nanning and Guilin substantially outperform peripheral cities. Critically, contrary to classical trickle-down theory, core-city growth has not generated significant spillover effects – a divergence attributable to karst topography constraints, homogeneous tourism products, and fragmented regional governance. Tourism endowment factors, particularly A-level tourist attractions, emerge as pivotal determinants of UT sustainability, while consumer price index exerts significant negative impacts with pronounced spatial heterogeneity. On this basis, we propose targeted strategies for UT sustainable development, including branded tourism agglomerations. This study contributes to deepening the understanding of UT disparities in economically underdeveloped regions and provides empirically grounded policy recommendations for fostering balanced and sustainable growth.