Environmental and Resource Economics 27: 109–133, 2004. © 2004 Kluwer Academic Publishers. Printed in the Netherlands. 109 Administrative Costs and Instrument Choice for Stochastic Non-point Source Pollutants ATHANASIOS KAMPAS 1 and BEN WHITE 2,∗ 1 Macaulay Land Use Research Institute, Aberdeen AB15, UK; 2 School of Agricultural and Resource Economics, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia ( ∗ Author for correspondence) Accepted 19 February 2003 Abstract. This paper extends the empirical analysis of non-point source pollution to the case where the pollutant is stochastic and alternative regulatory instruments have different administrative costs. It also applies a method of stochastic programming where emissions are log-normally distributed. For the Kennet catchment in South West England we rank a range of policies in terms of abatement costs alone, and total costs (abatement and administrative costs). On the basis of abatement costs alone, a uniform emission tax is the cost minimising policy, but on the basis of total cost a nitrogen input tax is the least-cost policy. Furthermore, the policy ranking, based on total costs, changes as the reliability standard increases. Key words: instrument choice, nitrate pollution, non-point source pollutants, stochastic programming JEL classification: H2, Q2, C61, R38, R52 1. Introduction The difficulties of regulating agricultural non-point source pollution are mainly due to the cost of information relating to monitoring and measurement (Braden and Segerson 1993). Typically agricultural pollutants are only detectable and measurable (if at all) after they have entered the ecosystem, thus identifying polluting sources is costly and may be impossible (Chambers and Quiggin 1996). Economists have proposed a range of theoretical approaches to overcome these problems. For instance Segerson (1988) proposes ambient taxes; Xepapadeas (1992a) proposes a scheme of random fines when producers collectively exceed the ambient standards; and Govindasamy et al. (1994) propose non-point tournaments. Reviews of the major issues of non-point pollution control can be found in Tomasi et al. (1994), Shortle et al. (1998) and in Shortle and Horan (2001). The issue of asymmetric information between the regulator (principal) and the farmers (agents) is by-passed in this paper following Griffin and Bromley’s (1982) reasonable assumption that the regulator knows the relationship between the farmers’ management practises and the released nitrate emissions, the so-called