A residential water demand equation is estimated using a panel data sample of five Portuguese local communities and 72 months, corresponding to a total number of 360 observations. Because of the presence of multi-part tariffs, we use the two common price-related variables: marginal price and difference as explanatory variables. To prevent the simultaneity bias from using observed quantities to determine the values of marginal price and difference directly from the rate schedule, we use an instrumental variable approach to create a constant marginal price and difference parameters for each rate structure.

The price elasticity values obtained fall within the range of those found in other case studies. Thus, although presenting weak elasticity, price seems to play a role in water demand management. However, we do not confirm the expected influence of difference on residential water demand. This can be a consequence of the complexity of the Portuguese water tariffs and the confusing signs that come from the simultaneous use of fixed quotas and increasing block tariffs. So, it is imperative to clarify water tariffs objectives by reviewing the Portuguese water tariffs design processes.

Author notes