The recent Sponge City Development (SCD) policy enacted in China has attracted tremendous investment, although its role in tackling water pollution remains vague. This study investigated the impact of SCD in Shenzhen, a national pilot city for SCD, on water quality in the transjurisdictional bay between Shenzhen and Hong Kong. A system modeling approach that coupled watershed nutrient load calculations with three-dimension bay water quality simulations was adopted. The main study findings include the following. While the planned SCD approach may result in much higher load reductions in total phosphorus and biochemical oxygen demand (BOD) and a comparable reduction in total nitrogen compared with those obtained by upgrading wastewater treatment plants, the contribution from green infrastructure (23.2%, 21.1% and 17.4% for total nitrogen, total phosphorus and BOD, respectively), which is a definitive component of SCD, is much smaller than that of pipeline construction (i.e., a gray infrastructure component of SCD). The overall cost-effectiveness ranking is green infrastructure > pipeline > upgrading wastewater treatment plants. On average, SCD in Shenzhen could reduce the total nitrogen and total phosphorus concentrations in the bay by 2.4% and 20.2%, respectively, but increase the chlorophyll a concentration by 5.5% due to improved light conditions. This study represents one of the few attempts to quantitatively assess the water quality impact of the SCD policy at a regional scale. The study findings suggest that SCD emphasizing green infrastructure should be preferably adopted by small-to medium-sized cities, and SCD for an individual city may only have a local impact on water quality and needs to be harmonized with other policies within the regional context.