Among the policy alternatives for limiting fertiliser use, a tax on chemical fertiliser is administratively the easiest option. Its efficacy depends partly on the responsiveness of fertiliser demand to price changes. This paper reviews the empirical methodologies available for estimating the price elasticity of a derived demand and presents the results of several econometric models. The econometric analysis shows that the response of UK fertiliser demand to own‐price changes is greater than has been assumed on the basis of programming studies of arable farms. This result establishes a necessary but not a sufficient condition for an effective fertiliser tax. The full impact of a fertiliser tax over the longer term depends on the extent to which technical change is driven by trends in relative price