Agriculture, Ecosystems and Environment 86 (2001) 59–73 Comparative evaluation of pesticide risk indices for policy development and assessment in the United Kingdom Jackie Maud a,1 , Gareth Edwards-Jones b,∗ , Fraser Quin a a Natural Resource Economics Department, SAC, Edinburgh EH9 3JG, UK b School of Agricultural and Forest Sciences, University of Wales, Bangor, Gwynedd, Wales LL57 2UW, UK Received 17 August 1999; received in revised form 15 August 2000; accepted 30 August 2000 Abstract The objective of this work was to review the appropriateness of five pesticide risk indices for use in policy development and assessment in the UK. A common data set of 133 pesticides, constituting 25 insecticides, three molluscicides, 40 fungicides, 60 herbicides and five plant growth regulators was used to enable a comparative evaluation of the risk indices. There was poor correlation between the ranking of the 133 pesticides as given by each of the five indices when only toxicological data were used in the risk calculation. The correlation was improved when data on recommended application rates were incorporated alongside toxicological data in the risk calculations. There was wide variation in the products placed in the top 10 (most hazardous) and bottom 10 (least hazardous) of the respective rankings when risk was based only on toxicological data. All the indices showed a degree of clumping in their classification of the pesticides, with many pesticides being grouped together in one quartile of the indices’ potential range. The use of these indices in policy development and assessment is problematical. The lack of correlation in the rankings between the indices makes selection of any one particular index difficult, and clearly the results of any analysis undertaken will depend on the index selected. Whilst calculation and manipulation of the indices was straightforward, data collection was difficult for nearly all indices. This is a major constraint to their utilisation in policy. © 2001 Elsevier Science B.V. All rights reserved. Keywords: Pesticides; Risk indices; Policy evaluation; Pesticide risk; United Kingdom 1. Introduction There is currently much societal desire to reduce the environmental impacts of pesticides and to derive policies which will achieve this aim (Beaumont, 1993; Buttel et al., 1992; Donati et al., 1993; Falconer, 1998; ∗ Corresponding author. Tel.: +44-1248-382441; fax: +44-1248-354997. E-mail address: g.ejones@bangor.ac.uk (G. Edwards-Jones). 1 Present address: Environment Agency, National Centre for Eco- toxicology and Hazardous Substances, Evenlode House, Howbery Park, Wallingford, Oxon OX10 8BD, UK. Pretty and Howes, 1993). (The term “pesticides” is used throughout this paper in a generic sense, and as such refers to all classes of pesticide, namely in- secticides, herbicides, fungicides and molluscicides.) In order to inform any such policy it is necessary to summarise the risks presented by pesticides in a suc- cinct and manageable form (Levitan et al., 1995). This should enable predictions to be made of the likely costs and benefits of any policy change, and should also enable the monitoring of policy induced changes over time. Several indices of pesticide risk have been developed that seek to summarise the impacts of pes- ticides in exactly this manner (Gutsche, 1995; Higley 0167-8809/01/$ – see front matter © 2001 Elsevier Science B.V. All rights reserved. PII:S0167-8809(00)00258-9