Since the early 1980s, the phenomenon of privatization has quickly spread worldwide, changing the balance between the state and the market in favour of the latter. Its adoption questions and replaces the traditional roie of the state in providing and controlling certain public services. One formerly predominantly public service to be affected by privatization is piped water services, as it is commonly argued that private suppliers stimulate greater efficiencies and innovations than public

more »

... ers. Most of those writing on this subject tend to focus narrowly on comparisons of public and private water companies in an attempt to argue that one or the other is best. Alternatively, some concentrate on the policy process through which privatization found favour. However, the thesis takes a very different approach to the analysis of privatizing water supplies, contributing to an area that has attracted little attention: its theoretical context and its implications for democratic politics. The aim of the thesis is to concentrate on, and extend, the types of assumptions - efficiency and innovation - inherent in arguments for privatization, thus providing a wide-ranging theoretical context in which to locate the privatization of water services. After discussing at some length exactly what comprises privatization, the thesis examines the theoretical foundations from which the policy originates. With reference to two case studies of privatization - Britain and Wales, which privatized water services in 1989, and New Zealand, which has not fully privatized its water, but is increasingly favouring more commercial practices - the thesis then illustrates how the theories have informed privatization in practice. The thesis concludes that privatizing water supplies . is an inappropriate extension of these theories because, first, they do not recognize the inherently non-commercial nature of water services and, second, because their implications for citizenship, and therefore democratic politics, are potentially very damaging.