In this paper, we investigate productivity growth in 55 major Australian urban water utilities using nonparametric frontier techniques over the period 2005/06 to 2008/09. The five outputs included in the analysis are chemical and microbiological compliance, and the inverses of real losses per connection, the number of water main breaks per 100 km of water main, and water quality and service complaints per 1,000 properties. The input is operating expenditure. Using Malmquist indices, we decompose productivity growth into technical efficiency and technological change. The results indicate that annual productivity growth averaged 1.04 percent across all utilities and was largely attributable to efficiency gain, roughly equally split between pure technical efficiency and gains from scale. As in many other highly regulated Australian industries, technological improvements during this period are very small, averaging just 0.22 percent a year. We subsequently employ second-stage regression analysis to quantify the effects of uncontrollable (nondiscretionary) factors on total factor productivity and efficiency change, including the role of different water sources (surface, ground, recycled and bulk water purchases) and utility size. The results indicate that imposed environmental factors only account for a small percentage of the observed variation in efficiency, technology change, and productivity improvements.