This paper is concerned with the timing of an agricultural policy reform under uncertainty. The focus is on the opportunity cost of giving up the option to wait when implementing a policy reform. Including the option value in applied policy analysis can help explain why conventional analyses may find observed policies to be Pareto-inferior. Furthermore, it explains why otherwise profitable policy reforms may be delayed. The theoretical model is applied to Norwegian agricultural policy anticipating a prospective WTO agreement. It is argued that the option value should be incorporated into applied policy analysis when high uncertainty prevails.