The impact of reforming municipal water tariff on water rationing and consumer welfare in Saudi Arabia

In Saudi Arabia, the non renewable groundwater reserves were estimated to be 259.1–760.6 billion cubic meters, with an effective annual recharge of 886 million cubic meters. The total renewable water was 2.4 billion cubic meters/year. About 1.4 billion cubic meters/year of runoff is stored by 302 dams, from which 992.7 million cubic meters is recharged to the shallow aquifers, 303.5 million cubic meters is used for drinking and 51.5 million cubic meters is for agriculture. The country produces about 1.06 billion cubic meters desalinated water annually, which is blended with groundwater for domestic water supplies (Chowdhury & Al-Zahrani 2012).

Saudi Arabia is considered to be the largest desalinated water producer in the world. Supplying municipal water for Saudi Arabia, with adequate quality, quantity, and timing, to satisfy the country's needs is absolutely necessary through an appropriate water strategy. Total annual municipal water use in Saudi Arabia has been estimated at 2.3 cubic kilometers in 2010, or 13% of the total water use. The price paid by individual households for municipal water is not based on the interaction of supply and demand, because water is price inelastic and has no substitutes. Under the current tariff system, the water price is 0.10 riyals for consumption in the 1–50 cubic meters block, 0.15 riyals for consumption in the 51–100 cubic meters block, 2.00 riyals for consumption in the 101–200 cubic meters block, 4.00 riyals for consumption in the 201–300 cubic meters block, and 6 riyals for consumption exceeding 300 cubic meters per month. According to the current tariff system, water is highly subsidized in Saudi Arabia (92% of the cost of supplying water), as water is the product of very expensive desalination. Both population growth and greater consumption are putting stress on water. It is tentatively estimated that the average daily water consumption for those connected to the network is about 235 liters per capita, a level lower than that of the United States. In order to conserve water and to guarantee drinking water for the current and coming generations, the government is contemplating raising prices by reducing its subsidies on water. Saudi regulators are discussing the structure of the tariffs, as 124.9 billion riyals in investments in desalination and water recycling plants are needed to meet the water demand (MWE 2011). Sustainable and efficient use of water requires the tariff to match not only the costs of production (i.e. the operation and maintenance (O&M) and capital costs), but also opportunity costs, economic externality costs, and environmental externality costs. Very often the tariffs do not even meet the full production costs.

Water conservation measures, such as awareness campaigns through the media and educational pamphlets, have been carried out. Higher water charges are considered as one of the most practical policies in order to cut down water consumption, and changes are critical for sustainability and to ensure viability of the national water sector. Since water is consumed both as a final good as well as an input good to various activities, then a clear understanding of the drivers of residential water demand is essential if water managers wish to craft effective demand management policies. Sustainable urban water pricing schemes must be designed to meet, amongst other requirements, the needs of current and future generations, efficient utilization criteria, full cost recovery (including production costs, opportunity costs and economic externalities), the economic viability of utility companies, and equity and fairness for different users (Klawitter 2003). Higher water rates allow utility companies to extend services to those currently not served, and can help maintain the sustainability of the resource itself through demand reduction, efficient resources reallocation, and increasing the supply, as well as improving equity, managerial efficiency, and the sustainability of the resource. Consumers and suppliers of water have different expectations of water tariffs. Consumers require high quality water at an affordable and stable price, while suppliers expect to cover all costs and have a stable revenue base. The level and structure of fees for water and wastewater services have consequences far beyond these expectations. Water-related fees can be expected to generate revenue, improve production efficiency, manage demand, facilitate economic development and improve public welfare and equity (Boland 1997).

The objectives of this study are as follows:

1. Estimating an aggregate price elasticity of demand for municipal water within the implemented increasing block system in Saudi Arabia.

2. Analyzing the welfare impact of raising the municipal water tariff.

Alhassan et al.’s (2016) study used population and agricultural trends to build futuristic water demand scenarios and proposed a set of variations in population growth and agricultural policy. It matched demand scenarios to variations by altering the contribution of desalination to explore the effect of policy options on groundwater withdrawals or as energy requirements for desalination.

Average monthly water utilized per household is as high as 72 cubic meters. Saudi households utilize significantly higher quantities of water compared to non-Saudi households. Also, household water consumption varies according to location, housing type, household size, and income (Al Kahtani 2009). The water pricing policy should be revised and adjusted to help conserve water. Of particular interest to water managers is the effect that a change in price will have on the quantity demanded. Without a proper understanding of how consumers respond to price, policy makers must rely on trial and error to achieve revenue and conservation goals established by the utility companies (Brookshire et al. 2002). As a result, the price elasticity of demand is typically the variable of interest, serving as one measure of the responsiveness of consumers to changes in price.

Much of the recent literature on the demand for water has used long term water demand models to show the benefits of regulating water demand for the agricultural sector, since it constitutes more than 80% of the total water demand. However, it does not undermine the effect of domestic or industrial water demand, since such demand for water will soon increase due to the constant increase in the population growth rate (Hamdar & Tamer 2014). The study of Ouda (2013) showed that the current Saudi water tariff system is heavily subsidized, with people paying less than 5% of the water production cost. The subsidy will dramatically increase and will reach 6.472 million US$ by the year 2020. Previous studies focused on the impact of increasing block rates on demand. The rate structure is determined by the government, in an attempt to generate a stable stream of revenue while simultaneously promoting the conservation and equitable allocation of water among households. Within Saudi cities, the existing rate structure is the increasing block rate. Households face a constant price over the first block of units consumed, but pay a higher amount over the next block. An increasing block rate has the advantage of promoting conservation among individual water users (Cavanagh et al. 2002).

There are two ways to estimate/forecast water demand. The first, a water requirements approach, attempts to estimate the quantity of water needed for a variety of activities consistent with an average household or individual (Nieswiadomy 1992). The second attempts to model demand for water as a behavioral phenomenon. Traditionally, researchers have relied on regression analysis as the primary tool for estimating household water demand, to isolate the effect of a change in price on the quantity demanded while controlling for other variables. These estimates then serve as the focal point of scenario analysis (Gegax et al. 1998). Block rate pricing creates two potential problems. First, traditional economic theory focuses on the marginal price as the sole variable driving consumer demand. However, under block rate pricing structures, two consumers may face the same marginal price but different average prices regardless of whether there are fixed costs. This raises the question of which price should be used when estimating demand. Second, embedded in consumers’ choice about how much they wish to consume is a choice concerning the price they wish to pay. That is, consumers simultaneously choose both the quantity they wish to consume and the price they pay. Ignoring this may result in biased estimates of the price elasticity of demand.

Many earlier water demand studies ignored the presence of block rates altogether, using the average price of water as the sole indicator of price (Agthe & Billings 1980). However, such an approach ignores the role that marginal prices play in determining consumer demand (Howe & Linaweaver 1967). Including only the marginal price would ignore the income effect associated with the block rate structure. For these customers, the marginal price alone does not reflect differences in the price paid for each subsequent unit (Taylor 1975; Nordin 1976). A difference term should be included to account for the difference in total cost paid by the household, given the rate structure they faced versus the total cost that would be paid by the household if they faced a constant price (Bachrach & Vaughan 1994).

Some studies employed the average price as the more important determinant of consumption choice. Average price is the price signal obtained at least cost by consumers, and it reflects an expenditure effect that is not captured in the marginal price specifications (Foster & Beattie 1981; Arbués et al. 2003; Gaudin 2006). The increasing block tariff (IBT) structure provides different prices for two or more pre-specified consumption blocks. The price rises with each successive block. The utility company must decide on the number of blocks, the volume of water use associated with each block, and the price to be charged for each block when designing an IBT structure (Boland & Whittington 1998). IBT is a progressive tariff. This allows the utility company to provide a lifeline to the poor at below-cost rate, and charge higher prices for use beyond this minimum volume. The subsidy allows the poor access to water and sanitation and promotes public health. Under this system, poorer households get access to low-rate water since they possess fewer water-consuming appliances, and it allows for rich-to-poor subsidies and industrial-to-household subsidies as well.

Early attempts to resolve the endogeneity problem created by the simultaneous choice of price and quantity modeled demand in a manner consistent with Taylor (1975) and Nordin (1976), but employed statistical techniques such as two-stage least squares or instrumental variables to account for possible endogeneity. The work of Hewitt & Hanemann (1995) represents a significant advancement in the development of a behavioral model under block rate structures. They developed and applied a discrete continuous choice model of consumer behavior to address this problem more formally. The model is a probability statement in which each observation is treated as if it could actually have occurred at any kink or linear portion of the household's budget constraint. The probability statement for an individual observation is a sum of joint probability statements, one for each kink and linear block in the budget constraint. Each joint probability includes the probability of the continuous choice of quantity consumed and the conditional probability that consumption occurred at that kink or block, given the choice of quantity in their approach, the choice of how much water to consume is modeled as conditional on, and separate from, the decision about the block in which to consume. However, the approach is computationally intensive, and requires data at the household level which is often not available. As a result few studies have utilized this approach (Pint 1999; Cavanagh et al. 2002).

This study relied on available aggregated data (MWE 2011), and a survey study carried out by Al Kahtani (2009). However, water demand studies differ mainly in the type of data used and the treatment of the price variable (Pint 1999).

It is clear that using the available data (household average water consumed in each block tariff and the corresponding water unit cost) will yield a naive misleading result of positive relation between the quantity consumed and price. Indeed, we need the amount of water that will be consumed in each block and the price which is willing to pay this quantity. Therefore, the theoretical framework and the specifications of the water demand function within the frame of IBT applied in Saudi Arabia depend on calculating the average block price (ABP) and average block water quantity (ABQ), then estimating the demand function with a mathematical form that captures the relative change in water quantity to a relative change in water price.

Welfare analysis of the impact of raising municipal water tariff rates needs to determine an initial point for the household water quantity and cost, which can be done by calculating overall equilibrium. The water price elasticity is used to calculate the consumer surplus for the household at that base quantity and price (equilibrium consumption value divided by twice water price elasticity; see Appendix, available with the online version of this paper), so that the second objective of this study (the impact of raising municipal water tariff) can be fulfilled.

Based on official data, the average monthly household consumption of municipal water was calculated for various served cities (or regions) of Saudi Arabia. Table 1 demonstrates the variations in household consumption among served cities.

Dammam has the highest household consumption of municipal water (272.70 cubic meters/household/month). Asir, Riyadh, Khobar, Madinah, Yanbuo, Jeddah, Qaseem, Makkah, and Tayef follow with household consumption levels of 264.82, 259.14, 245.27, 229.63, 207.52, 198.01, 180.58, 178.40, and 131.78 cubic meters respectively. The calculated average household consumption amounted to about 222.68 cubic meters. Accordingly, Dammam, Asir, Riyadh, Khobar, and Madinah are considered above average while Yanbuo, Jeddah, Qaseem, Makkah, and Tayef are considered below average in household municipal water consumption. The variations in household water consumption can be attributed to many reasons. One of these reasons is the nature of the data available, which is somewhat misleading as it is based on a metering system of municipal water. This system involves the problem of aggregation among some households.

Table 2 shows the virtual quantity of water consumed by households and the corresponding price; it is virtually expected that the average price paid by consumers in the first block equals the official rate levied for the block. However, in the second block, where a household exceeds a water quantity of 50 cubic meters and is still less than 100 cubic meters, i.e. an average consumption of 75 cm, it is virtually paying 0.1167 SR per cubic meter.

Data presented in Table 2 show that 18.83% of all municipal water consumption was located in the first block tariff (0.10 SR/cm). The second increasing water tariff block contains 41.25% of municipal water consumption. Both blocks contain 60.08% of municipal water consumption.

The relationship between willingness to consume water in terms of average monthly water consumption per household, as estimated and presented in Table 2, and the weighted average price was econometrically estimated using the ordinary least square technique. It is quite clear from the table that 100% of consumers are willing to consume the first 50 cubic meters. Only 18.83% of consumers are not willing to consume any extra water (in the second block) at the rate of 0.15 SR, so their consumption level is limited in the range of the first block where the rate is only 0.10 SR per cubic meter. In other words the quantity of water the household is willing to consume at the rate of 0.15 SR per cubic meter is about 50 cubic meters. The average household water consumption at the second block (100%–18.17 = 81.17%) is 40.585 cubic meter.

The estimated price elasticity of municipal water demand in the kingdom of Saudi Arabia is –0.39. This means that a 10% increase in the municipal water tariff rate would lead to a 3.9% reduction in municipal water consumption.

Raising the water tariff will affect water rationing, the amount of subsidy for municipal water, and the consumer surplus. Consumer surplus is estimated based on the water equilibrium quantity.

Based on the data presented in Table 2, the equilibrium monthly water quantity was estimated at 115.485 cubic meters, and the municipal water base price was estimated at 0.5492 SR per cubic meter. The estimated municipal water price elasticity was used to calculate the consumer surplus for households, which was estimated at the base water quantity and price. The estimated consumer surplus was about 81.268 SR for households covered by water provision. The number of households in water service accounted for 869,869 households in 2009. It is expected to reach 963,222 households in 2014 by using the double exponential smoothing method. Accordingly, the consumer surplus was estimated at 70.693 million SR in 2009, and was expected to reach 78.28 million SR in 2014.

To simulate the impact of raising the municipal water price (i.e. point (2) to be compared with the initial point (1)), it is assumed that the price rate of the second and third block are raised by 33 and 50%, respectively. According to this assumption, the price rate would reach 0.20 SR for the second block, and 3.00 SR for the third block. Table 3 represents estimates of the weighted quantity of virtual municipal water consumed by households (ABQ) and the corresponding weighted average block price (ABP) at point (2) as a result of assuming a rise of municipal tariff in Saudi Arabia. Using demand function price elasticity and increasing the tariff rate by 33%, virtual water was estimated to be about 35 cm for block 2, and 32 cm for block 3. It is expected that the percentage of household consumption in each block will be affected as the assumed price rises, this can be captured using the new virtual water quantity relative to the block classes prices, and hence a new point for the overall quantity and price (ABQ and BPA) can be calculated. According to the results in Table 3, the shift in the demand function can be realized from estimating data.

The impact of a hypothetical increase in the municipal water tariff was predicted as follows:

1. The percentage of households in block (1) increased to about 29% (initial point 19%).

2. A decrease in the percentage of water consumers in the third block from 25.61 to 17.82%.

3. Relative stability in the percentage of water consumers of the second block around the existing percentage (38.48%).

4. An increase in average consumer payments for water consumed (price weighted average). The weighted price average would reach 0.1333 SR per cubic meter in the second block (while it was 0.1167 SR per cubic meter), and would reach 2.9 SR per cubic meter in the fifth block (while it was 2.61 SR per cubic meter).

5. The monthly equilibrium quantity would be 104.364 cubic meters, the average price would be 0.619 SR per cubic meter.

6. Average monthly saving in municipal water accounts for 11.121 cubic meters per household, which leads to a total annual saving of about 128.544 million cubic meters.

7. The percentage increase of the average equilibrium price would reach 12.7%.

8. The percentage decrease in consumer surplus per household accounts for 9.43% (which equals a decrease of 7.67 SR per household).

The main contribution of this study is to provide an easy method to estimate municipal water demand, using the percentage of household consumption of increasing block price rates. Municipal water demand price elasticity was estimated for Saudi Arabia by using the double log demand function. Therefore, water rationing and improving the water use efficiency in Saudi Arabia can be achieved with new price rates. The municipal water demand in Saudi Arabia was found to be of low price elasticity (–0.39%) at the average rate. The study strongly recommends increasing the recent water tariff rates by 33%. This will lead to total savings of about 128.544 million cubic meters in municipal water consumption. The reduction of the consumer surplus would not exceed 9.43%, which is very low compared to water losses and governmental subsidy.

The authors thank the Deanship Scientific Research at King Saud University for funding this research.