Too many water engineers have become accustomed to using the term 'demand' as a simple standard factor in estimating the necessary scale of a new facility.

When they are asked by their employers to design a new water supply system, or increase the capacity of an existing one, they first estimate the number of people to be served, for a domestic system, or the size of the area to be irrigated. These are then multiplied by 'unit demand' numbers based on long experience. They 'know' how many litres householders will take, or cubic metres that farmers will take, on average, if the water is effectively free.

To say, on this basis, that the demand for water at some point in the future will be xxx million m3/yr is very much like saying that the "demand" for new Mercedes in that year will be xxx million cars. It just might be if they were priced at $10 each! The truth is that we take huge quantities of water, and waste major amounts of it, simply because it is so inexpensive.

We would not thoughtlessly flush 20 litres of water down the toilet as often as we do now if the price were to rise to $1.00/litre. And we would not use 10,000 m3 of water to irrigate a hectare of rice in the dry Sacramento Valley of Northern California if it was going to cost us $1.00/m3.

In the future, all estimates of demand will need to be based on the cost of the water to be supplied.