Water-quality markets that allow point-nonpoint trades assume that nonpoint best management practices (BMPs) achieve the targeted reductions as soon as they are implemented. However, changes in water quality in response to BMPs occur over time—from a few months to decades. We simulate emission allocations using static and dynamic-optimization models to determine whether a simple static allocation can produce results comparable economically and environmentally to complex multi-period designs for nitrogen emissions to Chesapeake Bay. We find that static rules provide relatively large cost savings compared to dynamic rules but result in a delay in achievement of water-quality targets.