late 1980's and early 1990's and continue to work after the drought conditions have ceased.
Conservation programs may be ultimately necessary simply to counteract an exogenous long-
term increase in residential use.
The climate affects residential use in predictable ways. Water use is strongly correlated
with average monthly temperature and seasonal variation in temperature. Precipitation was
consistently insignificant in all models. All cities in this analysis are semi-arid to arid in climate
and thus the ratio of evapotranspiration to precipitation is much greater than one. Landscape
watering is necessary if one wants to maintain traditional residential lawns and trees. Random
and infrequent rains do not change residential watering patterns to a significant degree. Other
variables, exogenous to a water utility, such as residential income and the size of the city have a
relative minor impact on residential use.
We use the empirical demand curve results from the Michelsen study by applying the
function to the climatic and socioeconomic conditions of Albuquerque, Las Cruces, Santa Fe and
El Paso. The Michelsen model has a demand elasticity of -0.30. The demand model is
remarkably effective in predicting water use in the four cities. The law of demand implies that
people will use water up to a level where the value of an additional unit equals the cost of
purchase. In Albuquerque this is about $1.59 per 1000 gallons. This water price is known as the
marginal price. Since Albuquerque uses an increasing block rate structure, this is the price that
applies to purchasing the next 1000 gallons for an average residential consumption level.
But what were to happen if future water supplies do not keep up with population growth?
The amount of water use per household would have to be reduced. Table 12 outlines this
potential situation for Santa Fe, Albuquerque, Las Cruces and El Paso. As water use is cut back
people are willing to pay a higher value per unit, but are unable to obtain it. These unfilled