The subject matter of this thesis is the definition, measurement and use of marginal cost as a tool of analysis to assist the process of decision-making in the water supply industry. Demand management is viewed in broad terms to include the establishment of an optimal structure and level of prices and investment in optimal capacity as well as investment in demand-restraining measures such as leakage detection and control. The study examines the definition of marginal cost as a benchmark for price setting. It provides empirical estimates of the various components of marginal cost of water supply in the Hampshire area, part of the Southern Water Authority. These estimates assume an exogenously determined level of demand and therefore exclude any possible direct interaction between the pricing and investment decisions. Departing from this tradition the study also examines a number of models where, under specific assumptions, optimal prices, output and capacity levels over a chosen planning horizon are simultaneously determined. This allows for direct interaction between the pricing and investment decisions. The study simulates optimal paths of prices, output and capacity expansion in the Hampshire area. This is carried out under various assumptions, one of which admits the potential of staging capacity expansion in order to take advantage of economies of scale in the capital cost function. An analysis of leakage detection and control as a demand management tool is presented in the final part of the study. The purpose of this analysis is to investigate how leakage detection and control may be conducted using either cost-benefit analysis or an appropriately defined tool of marginal cost.