This study examines the impact of agricultural input prices on river water quality in China, using provincial panel data and high-frequency monitoring records for chemical oxygen demand (COD) and ammonia nitrogen (NH₃-N). In our own research, we have collected data from 30 provinces from 2010 to 2020.Employing a two-way fixed effects model, we find significant heterogeneity across input types. Rising fertilizer prices reduce pollution through cost constraints and input substitution, supporting a price-induced reduction effect. Pesticide prices show no statistically significant effect, reflecting their inelastic demand as essential crop protection inputs. In contrast, higher mechanization costs are associated with increased pollution, as they discourage the use of efficient machinery and prompt a return to less precise traditional practices, thereby exacerbating nutrient runoff. These results suggest that price-based policies are not uniformly effective. We therefore propose a three-pronged strategy: reforming fertilizer pricing to curb overuse, providing targeted subsidies for green mechanization to lower adoption barriers, and establishing cross-regional governance frameworks to address spatial spillovers in water quality management.