This article analyses the relation between mineral rents and development outcomes at the subnational level. The classical literature suggests that natural resource abundance has negative effects on well-being, a situation referred to as the resource curse. However, a novel strand of research emphasizes that rentier states worldwide exhibit contrasting outcomes. To account for such variation, this investigation aligns with approaches stressing the significance of contextual (place and institutional) factors to studying the resource curse. The main claim in this work is that both structural and institutional factors related to the extractive industry help account for variation in development outcomes. It contends that mineral rents are positively associated with human development and economic industrialization when the extractive sector is not geographically concentrated in enclave economies, and subnational fiscal institutions redistribute enough rents from producing to non-producing districts. It empirically tests this argument using a time series cross-sectional analysis, a difference-in-difference (DiD) estimation, and two case studies in Argentina, a country where subnational territorial units collect mineral royalties and have exogenously created their own rent-sharing regimes. It finally provides some comparative implications that may contribute to current debates on the socioeconomic impact of natural resource wealth.