OLS and Instrumental Variable Price Elasticity Estimates for Water in Mixed-Effects Model Under Multiple Tariff Structure

Abstract

A mixed-effects residential demand model for potable water is developed using a longitudinal data set constructed for the analysis. The data set comprises of 1,065 households from the Sydney Metropolitan and Wollongong areas, covering sixteen quarters from 1990 to 1994. The purpose of developing the demand model is to use it as a base model to forecast water demand changes in response to changes in the tariff structure. The empirical results show that consumers do respond to the marginal price while faced with the multipart tariff structure. Therefore price can be considered as an influential tool in the implementation of demand management strategies. However the magnitude of price elasticity suggests that substantial increases in price would be required to influence demand. OLS and Instrumental Variable\Maximum Likelihood estimation techniques are employed to conduct the analysis. The results support both - theory and past research, which states that IV/ML estimation technique tends to produce unbiased and consistent estimates than OLS, when price depends on quantity consumed. The Taylor/Nordin theory is also tested and the results are supportive of the theory.