An Estimation of Residential Water Demand Using Co-integration and Error Correction Techniques

The purpose of this paper is to measure the short- and long-run effect ofthe price of water on residential water use. Unit root tests reveal that water use series and series of other variables affecting use are non-stationary. However, a long-run co-integrating relationship is found in the demand model, which makes possible to obtain a partial correction term and to estimate an error correction model. The empirical application uses monthly time-series observations from Seville (Spain). The price-elasticity of demand is estimated as around -0.1 in the short run and -0.5 in the long run. These results are robust to the use of different specifications.