Rural tourism sustainability depends on economic returns, community participation, cultural continuity, and ecological stewardship. This study uses a resource exchange perspective to explain how multiple actors co-create value in rural tourism and how institutional coordination and service interaction transform resources. Drawing on more than 90 days of fieldwork in Zhushan Village, western Hunan, China, between July 2021 and July 2025, the case study uses 96 semi-structured interviews, participant observation, and supplementary sources. Qualitative content analysis combined deductive and inductive procedures. The findings show that actors form exchange relationships around heterogeneous resources, including capital, land, dwellings, skills, knowledge, information, affect, reputation, and institutional support. Resource complementarity and cost–benefit assessments shape relationship formation, while reciprocity and fairness sustain cooperation. Institutional coordination stabilises resource commitment through arrangements for participation, rights, responsibilities, and benefit allocation; service interaction expands resource use through joint practice. New uses and related claims for participation and benefits then prompt institutional adjustment. Feedback between the two mechanisms connects dispersed resources in tourism operations and collaboration over public affairs, producing economic, social, cultural, emotional, and ecological value manifestations.

This study examines county-level correspondence between rural labor out-migration and cultivated-land adjustment across 122 units in Hunan Province, China. The rural labor–land correspondence (RLLC) indicator uses positive net cultivated-land reduction between benchmark years, not registered farmland transfer. With 2005 as the baseline, RLLC was calculated for 2010, 2015, 2020 and 2023 and analyzed using spatial autocorrelation, ordinary least squares, geographically weighted regression and geographically and temporally weighted regression. RLLC exhibited spatial unevenness and positive autocorrelation (Moran’s I = 0.3777–0.5250). Among predictors without direct overlap with RLLC components, household register urbanization had positive mean coefficients in most counties but limited local statistical support. Associations for agricultural business entities, grain-production specialization and per capita GDP were positive and geographically localized where they exceeded the reporting threshold, while per capita fiscal expenditure had negative mean coefficients throughout. In the OLS sensitivity analysis excluding three component-overlapping predictors, the five remaining predictors retained their coefficient directions and statistical significance. RLLC does not measure farmland transfer or optimal land allocation, and the observed associations do not establish causal effects. Farmland transfer is considered a possible institutional mechanism.