Français      print      e-mail    

Abstracts

After a three-year period of drought in France, many debates have been held on water supplies. The price of public water utilities is considered as means to improve solutions of quantitative and even qualitative problems of water supplies. Surprisingly, only one study has been published in France on the real influence of public water utility price on domestic water consumption.

Different evaluations of domestic water consumption price elasticity are presented. They use the (price, consumption) data from a national panel of 500 public water utilities for the years 1975, 1980, 1985 and 1990, and front an even larger sample of 1362 public water utilities for the years 1975 and 1980.

Domestic water consumption for each utility is computed as the volume of water sold to domestic users (industrial and agriculture consumption are ignored) divided by the number of permanent equivalent inhabitants. The number of permanent equivalent inhabitants is the total of bath permanent inhabitants and seasonal inhabitants weighted with the annual duration of their stay.

The measure of the price for the public water utilities used is the price that a domestic user pays for an annual consumption of 100 cubic meters, free of taxes. This price does net include the price paid for public sewerage.

The evaluations of elasticity are carried out in two different ways : time series and cross-sectional. In the time series approach, the evolution of the domestic water consumption for the utilities which have raised their price sharply between one year of observation and the next is compared with that of utilities whose price has remained constant (in constant value) during the same period. For each of the three periods considered (1975-1980, 1980-1985 and 1985-1990), the evolution of the consumption for the utilities that have raised their price appears to be lower. Values of short-term elasticity between - 0,11 and - 0,17 can be derived from this statement. The evolution up to 1990 of the consumption of the utilities that had raised their price between 1975 and 1980 shows that the total effect on consumption of a price increase can take 5 to 10 years to be completed. A long-term price elasticity of -0,33 can then be calculated. The tact that the price for public sewerage is not taken into account is assumed to be a source of small under-estimation for the results of elasticity that are found.

In the cross-sectional approach, thanks to the important quantity of data (price, consumption) that are available for 1975 and 1980 (1336 data points for each year), it is possible to show a relation between domestic water consumption and price for different utilities for bath 1975 and 1980 without any additional information factors other than price that can influence water consumption. The method used consists of dividing the 1336 utilities into 8 groups of services whose prices are similar. For each group, the average price and average consumption is computed. The important number of utilities and users that each group represents minimizes the influence of factors other than price that can influence domestic water consumption. The influence of the price paid for public sewerage is globally taken into account for each group of services through the proportion of users who benefit from public sewerage. Public sewerage price is considered to represent 45 % of the group average price for water utilities. The values of elasticity that can be deduced, which are between short-and long-term elasticity, vary from - 0,23 to - 0,29.

The hypothesis of rational consumer behaviour implies that consumers should react to the marginal price at their own level of consumption, rather than to the average price. However in France, marginal price and average price for one utility are very close, due to the tariff structure and, in tact, consumers probably react as much to average price as to marginal price.

Taxes that can overcharge consumers bill were unknown - it was impossible to take them into account - but they are judged to be low enough not to change the values of elasticity that were found.

The different methods for the calculation of price elasticity, which are presented here for several series of data, give results which are close to each other. Comparison with results from other studios that are available, especially those from the United States, is difficult because the dispersion of the values found is large. Nevertheless our results can be judged rather close to most of the other comparable results found abroad.

Keywords

Public water utilities, price of water, water consumption, price elasticity.

Corresponding author

Boistard, P., Ecole Nationale du Génie Rural, des Eaux et des Forêts, ENPC-CERGRENE, 19, Avenue du Maine, 75015 PARIS

Français      print      e-mail