Water-quality trading: Can we get the prices of pollution right?

Water‐quality trading requires inducing permit prices that account properly for spatially explicit damage relationships. We compare recent work by Hung and Shaw (2005) and Farrow et al. (2005) for river systems exhibiting branching and nonlinear damages. The Hung‐Shaw scheme is robust to nonlinear damages, but not to hot spots occurring at the confluence of two branches. The Farrow et al. (2005) scheme is robust to branching, but not to nonlinear damages. We also compare the two schemes to each other. Neither dominates from a welfare perspective, but the comparison appears to tilt in favor of the Farrow et al. scheme. Key Points: Water‐quality trading is a useful policy approach to protecting water quality Trading is hard, especially for branching rivers in which damages are nonlinear Simulations show nonlinearity is the lesser problem, emissions limits critical