This paper discusses the rose of information structure (i.e., information cost, reliability, and distribution among agents) in the design of a regulatory mechanism for controlling non-point source pollution. An ambient concentration tax mechanism is examined for non-point source pollution with spatial transport among multiple zones. Imposition of the tax requires costly measurement of ambient concentrations in selected zones, and the selection of zones for measurement must be undertaken without perfect information regarding several parameters of the problem.

Potentially crucial information issues in this context include: (a) the cost of measuring ambient concentration may exceed the benefits from imposing the tax, though these costs can be substantially reduced by carefully targeting monitoring sites; (b) producers' responses to the tax depend on prior beliefs regarding the fate and transport mechanism, and the efficacy of the ambient tax will depend upon the regulator's ability to ascertain and, perhaps, modify those beliefs; and (c) without regard to the extent to which the ambient concentration tax is imposed, it may be optimal for the regulator to acquire additional information regarding the fate and transport mechanism, either for the entire region or for specified "problem" zones.