Impact of water quality regulation on the agricultural economy in China

Agricultural production is a major contributor to water pollution in China and increasingly subject to regulation. However, the effectiveness of such regulation—particularly for non-point source pollution—remains underexplored. This paper estimates the causal effects of environmental enforcement on agricultural outcomes using a spatial regression discontinuity design based on China's surface water quality monitoring network. Monitoring stations only register pollution from upstream sources, creating sharp spatial variation in regulatory pressure. After water quality performance was linked to political promotions in 2003, upstream counties experienced a significant decline in agricultural value added—about 58% lower than downstream neighbors. No such gap existed before the reform. The effect is robust to placebo tests, bandwidth changes, a difference-in-discontinuities design, and an instrumental variables strategy using hydrological station placement. To explore the channels through which regulation affects agricultural outcomes, we combine administrative data, household survey microdata, remote sensing, and emissions inventories. We find that upstream counties hired fewer agricultural workers, cultivated less land, and applied less fertilizer. Rural households in upstream counties are also more likely to migrate across counties, consistent with declining local agricultural viability. Remote sensing data confirm broad reductions in crop planting areas. N 2 O emission inventory data show accompanying reductions in N 2 O from both agricultural soils and manure. These results show that China water quality regulation can alter input use, land allocation, and emissions in agriculture. However, enforcement may also displace rural labor and reshape household decisions, highlighting important trade-offs in implementing environmental policy in developing country contexts.