APPENDIX 8: COUNTRY CASE STUDY - WATER POLICY REFORM IN SAUDI ARABIA

ACRONYMS USED IN THE TEXT

1. INTRODUCTION

Water availability is essential to sustainable economic and social development in any country. In an arid country with no perennial rivers, such as Saudi Arabia, groundwater is the main, but limited, resource that must be used wisely in industrial, residential and agricultural areas to guarantee an adequate supply of water in the future. Current policies, however, encourage the use of groundwater supplies without regard for the fact that withdrawals substantially exceed natural recharge. Policies are socially non-optimal that artificially encourage groundwater withdrawals for agriculture at the same time as the country spends thousands of millions on desalination plants to augment municipal and industrial water supplies. Similarly, policies that provide costly desalinated seawater to municipalities at highly subsidized rates that do not reflect its social cost encourages the misuse of the resource. Government policy-makers, particularly those in the Ministry of Planning (MOP) and the Ministry of Agriculture and Water (MAW), as well as independent researchers, are becoming increasingly concerned with the rate at which aquifers are being depleted, the implications of that for future water security for Saudi Arabia, and its economic cost. It is to these concerns that this paper is addressed.

Figure 1 Wheat production, area and prices in Saudi Arabia

(Sources: MAW and IFPRI data)

The paper is in great part motivated by the structural development that led to the changes indicated in Figure 1. Wheat output has increased by about 26 fold; high prices and heavy investment increased output from virtually nothing prior to 1980 to about 4 million tons by 1992. As a result, net imports of 500 000 t in 1980 were converted to net exports of more than 2 000 000 t in 1992. Farmers are paid approximately three times the world market price for wheat. Inputs such as fertilizer and equipment are provided at substantially less than their world market prices. Subsidized credit is available through the Saudi Arabian Agricultural Bank (SAAB). The result of favourable government policies has been to create a boom in agriculture, the sector growing at 12% per annum during 1979-91.

However, increased agricultural output has been accomplished at a high budgetary cost; more important, it has been achieved only by using groundwater resources that are mostly non-renewable. Because agriculture is based entirely on groundwater, the attractiveness of investment in the agricultural sector has been accompanied by a significant fall in the region’s water table. The effect of this fall in the medium term is increased pumping costs, along with a decreasing aquifer-specific yield. Moreover, the use of groundwater to grow highly subsidized agricultural commodities has occurred at the same time as the government has been spending thousands of millions on desalination plants to meet municipal and industrial demands for water, indicating a conflict of objectives between agricultural policy and any meaningful long-term water policy, because of the competing demand for the limited water resources between agriculture on the one hand, and municipal and industrial use on the other. Water consumption in agriculture constitutes more than 90% of total water use. The amount of water used in agriculture is inextricably linked to agricultural policies (Al-Sheikh, 1994).

In analyzing the interaction between agricultural policy and the optimal allocation of water reserves, the focus is on the economic aspects and trade-offs between current and future policies and objectives. This study pursues several objectives: (1) to assess the supply and demand of water in Saudi Arabia; (2) to understand more about the economics of water use in the principal sectors; (3) to evaluate the impact of government polices on optimal water use; and (4) to analyse the optional allocation of water over time, space, sector and source at both private and social prices.

The study is divided into six sections. Section 1 introduces the paper. Section 2 outlines the geological and climatic features of Saudi Arabia. Section 3 describes the supply of water resources in Saudi Arabia. Section 4 assesses the water demand in both the agricultural sector and the municipal and industrial sector. In addition, special attention is given to the costs of desalinated water to both government and consumers. Section 5 analyses the agricultural policy of Saudi Arabia and the policy instruments that the government uses to influence incentives, production and returns. The section also provides some background on property rights. Finally, Section 6 summarizes the policy recommendations and conclusions reached. In particular, the current wheat policy is analysed in the context of its objectives and water consumption.

2. GEOLOGY AND CLIMATE

Saudi Arabia is located in the furthermost part of southwestern Asia. Its land constitutes about four-fifths of the Arabian Peninsula, covering approximately 2.25 million km² (about one-fourth the size of the United States of America). Precipitation is scarce and infrequent, and, as a result, almost the entire country depends solely on groundwater to irrigate wheat, barley, coarse grains, vegetables, fodder, dates, fruits and other perennial crops and plants. Annual rainfall averages 100 mm, but when rainfall is heavy, it produces a considerable run-off (Authman, 1983). The land of Saudi Arabia is of varied topographic structure that is largely desert. Although it has many large valleys that slope from the Sarawat Mountains in the south, no permanent rivers or streams exist in the area. The climate for the most part is harsh dry desert with great extremes of temperature; the exceptions are the coastal strips and southwestern mountains. The summers for most of the country are usually long, hot and dry, while the winters are short and cool, with large differences between the daily minimum and maximum temperatures and between summer and winter temperatures. In some places, summer temperatures exceed 49°C (120°F) during the day.

Humidity near the sea reaches about 90%, but for most of the desert inland it is less than 10%. In fact, aridity is the dominant climatic feature of Saudi Arabia. Not surprisingly, the evaporation level is high, but differs from one location to another. For example, in Al Kharj the evaporation level is 3 753 mm during the year, while in Al Zilfi it is as high as 5 439 mm. In general, evaporation levels range from 400 mm to 4 500 mm during summer alone. This high level is the result of high temperatures, high winds and low levels of rainfall. The high temperatures, high evaporation levels, low humidity and infrequent rainfall mean that crop water requirements for irrigation schemes are very large. For example, wheat grown from November to March requires approximately 13 173 m³/ha of irrigation water.

3. WATER SUPPLY

Like other commodities, the economics of water resources in Saudi Arabia depends on supply and demand. Water supply in Saudi Arabia comes from eight principal aquifers and nine secondary aquifers, and from limited surface water, desalination and reclaimed wastewater. The distinction between the principal and secondary aquifers is based on their hydrologic properties and areal extent. Principal aquifers have greater permeability and larger yields than secondary aquifers. However, recharge is small in all aquifers, about 1 270×10⁶m³/year (MOP, 1985).

This section’s main focus is on the supply of, rather than demand for, water. Saudi Arabia sits on aquifers, some very deep, that contain large amounts of fossil water. These aquifers are spatially disperse, with very different pumping characteristics according to the particular geology of the areas. Unlike other aquifers, little of the water pumped returns to the aquifer. Pumping from the deep aquifers, in particular, amounts to mining the resource.

Information about groundwater supplies in Saudi Arabia is rather old. The most recent account of groundwater supplies in the Kingdom is the Water Atlas of Saudi Arabia (MAW, 1984). Since that date, major changes in water demand have taken place, significantly altering the water data. It is important also to note that the storage parameters for aquifers given in the Water Atlas and in the Fourth Development Plan (1985) were based on hydrologic studies done in 1966, 1969 and 1980 (MAW, 1984: 47). The Fifth Development Plan (1990) recognized the deficiency of the data and called for more reliable data about water resources and for the completion of the National Water Plan. Five years later, the Sixth Development Plan (1995) called for the same objectives, indicating that no hydrological studies had been accomplished during that period. It was not until early 1996 that the Ministerial Council issued a decree outlining the objectives and phases to complete the national water plan.

3.1. GROUNDWATER RESOURCES

Principal aquifers of the country contain large amounts of fossil water but have little recharge. The principal aquifers are Wasia-Biyadh, Wajid, Um Er Radhum, Minjur-Dhurma, Saq, Tabuk, and Dammam-Neogene. Some of these aquifer catchment areas extend across national borders. Saq and Tabuk aquifers extend across the national border to Jordan and Syria. Dammam, Neogene and Umm Er Radhuma aquifers extend to Bahrain, Iraq, Kuwait, Oman, Qatar and the United Arab Emirates, while Wajid aquifer extends into Yemen. Currently, however, Saudi Arabia is the major user of these aquifers. Within Saudi Arabia, most of the known groundwater resources have already been tapped. Ancient deep aquifers as well as shallow aquifers are being mined, meaning that the present extraction rate is higher than its recharge rate. Thus, future generations will be left with less water resources for their increasing populations.

For example, in the Riyadh region, the Minjur aquifers form the first of these deep groundwater resources. The Minjur formation starts at about 1 200 to 1 500 m below the surface. Its water quality differs in terms of location and depth, ranging in total dissolved solids (TDS) from 1 200 to 15 000 mg/l. Specifically, chloride and sulphate concentrations increase with water levels that rise from 200 m below the land surface in Riyadh City to 10 m above the land surface in the Kharj area. The water levels in Riyadh City, however, declined after the 1950s due to an increase in withdrawals, causing the piezometric level (the level to which water from a confined aquifer will rise in a well) to fall from 45 m in 1956 to 170 m below surface in 1980 (MAW, 1982).

The Dhruma Aquifer consists of shale and some sandstone that actually becomes part of the Minjur Aquifer. The maximum thickness of the aquifer reaches 375 m, and its minimum is 264 m. Water in this formation is used for agriculture in the Zulfi Plain, Wadi Birk and the Dhruma Plain. In Zulfi, the water level dropped by about 6 m between 1968 and 1978 due to mining (El Katib, 1980).

Secondary aquifers exist in various areas. The main aquifers in this category are Khuf, Tuwail, Aruma, Jauf, Sakaka and Jilh, in basalt and alluvial areas. These water systems have been tapped for years. For example, in the Riyadh region, Wadi Hanifah is one of the oldest centres of farming in the Tuwayq mountain area. Groundwater has been tapped using motor-driven pumps since the 1940s, at an average extraction rate of 28×10⁶m³/year. Other data indicate that, by the late 1950s, the average yearly recharge in the area was much less, at approximately 7.36×10⁶m³, while the annual extraction rate was still very high, about three times as much (Sogreah, 1969). The general idea is clear: water is being used much faster than it can be replenished.

A second shallow alluvial water system in the Riyadh region is the Jubaylah aquifer. This aquifer is about 100-150 m thick, and many of the wells that tap it, including city wells, have closed due to falling water levels. More than 80 wells, however, have been in continuous operation since 1950. The result is not unexpected. Even in 1967, average daily production of 30 700 m³ was greater than the recharge of 29 000 m³ (Sogreah, 1968). Hence, Jubaylah water levels declined, and the Riyadh City Authority, a large user of this aquifer, began nearly 20 years ago to use deeper aquifers (e.g., Minjur) and more distant aquifers (e.g., Hair, Wadi Nisah and Wasia) to meet city water requirements. The city water authority also started using desalinated seawater transported from Jubail through pipelines across 470 km of desert to Riyadh.

Table 1 Groundwater resources in Saudi Arabia

Sources: 1984 data - MOP, Fourth Development Plan, 1985: 134. 1996 data - estimates.

Table 1 summarizes the groundwater resources in Saudi Arabia. The data show the approximate quantities of water stored in the main aquifers of the area. The principal and secondary groundwater aquifers storage amounts to 500 000×10⁶m³. However, as mentioned earlier, these are from data published in 1984. Furthermore, the hydrologic studies were done earlier than that and the storage coefficients have never been updated to reflect the volume withdrawn from storage during that period. MOP estimates for resources remaining, based on conservative extraction rates, is 289 130×10⁶m³, reflecting about a 43% decline over the past twelve years in the country’s total non-replenishable water stock, a stock formed over thousands of years.

MAW monitors groundwater levels, with over 761 observation wells in the country (MFNE, 1990). These wells measure water levels on a continuous or periodic basis. Water levels usually do not rise or fall uniformly throughout an aquifer or a system of aquifers when a charge or a discharge occurs. Similarly, changes in a single well do not indicate a uniform change in the water level of the aquifer. Thus, the fact that water level in a specific location of an aquifer is declining does not imply that an overdraft problem is occurring. However, when a network of observation wells or cross-regional observations indicate a decline in water levels, then the trend of groundwater is on the decline, even though at specific locations water levels can be static or increasing. As indicated in the discussion earlier about groundwater resources, water levels in specific aquifers have been declining. So, not surprisingly, the MAW’s observations, both regionally and nationally, indicate that water levels in wells have declined in all types of aquifers. For example, according to the Saudi Arabian Water Atlas (MAW, 1984), the water level in the Wasia aquifer declined at well M1-1-W from +131 m above the land surface in 1978 to +125 m above land surface in 1982, a net decline of 6 m in water level during a period of merely four years. Furthermore, in Wadi Hanifah near Riyadh City, the water level declined 2 m per year. In addition, the water level in the Minjur Aquifer dropped by 85 m at well M-2 in Buwaib wellfield between 1978 and 1983 at an average rate of 7.5 m/year. Likewise, water levels of wells tapping shallow aquifers have declined by 2 m or more per year throughout the Riyadh region. In sum, one can generalize that the average decline in water level is 2 to 3 m/year (MAW, 1984).

Data from fieldwork conducted in 1991, collected from drilling companies as well as farmers, covering water level and well depth at various locations in the Riyadh region, showed a similar decline. As expected, water level and well depth are location- and formation-specific. Also, as mentioned earlier, water levels had fallen in all of the study areas by an average of 2.1 m/year, indicating that groundwater resources are being mined. Whereas the average static water level in 1980 was at 64.9 m below surface, in 1990 the average static water level had fallen to 92.3 m below surface. Thus a deepening of water pumps by 2 pipes (6.4 m) every three years on average is required (Al-Sheikh, 1994). This leads to an increase in the marginal cost of extraction per cubic metre of water. First, expenditures are higher for drilling, casings and other related costs. Second, water quality, in terms of the higher temperatures and greater salinity of water from deeper water formations, makes such water unsuitable for certain agricultural purposes, such as fruits and livestock. For example, in Minjur temperatures can reach 65-80°C, requiring tapped water to be cooled prior to application.

3.2. DESALINATED WATER

Until the 1930s, water supplies to Riyadh City were based on hand-dug wells. Then water levels began to fall and the quality of water also started to deteriorate due to seepage from sewage. The High Commission of Riyadh Development estimates that seepage is occurring in the neighbourhoods of the capital at a rate of 560 000 m³/day. The water distribution network leaks 162 000 m³/day, accounting for 29% of the total seepage, which constitutes about 17% of the water imported to the city. The rest of the seepage comes from landscaping, gardens, sewerage systems and septic tank lines. This seepage leads to serious engineering, environmental and health problems in Riyadh City (Alymamah, No. 1160, 14/12/1411H). In 1956, the government started drilling deep wells and tapping both the Minjur Aquifer and areas that were distant from the city to provide an abundant supply and good quality of water. In 1963, the Biyadh and Minjur formations were used. The water from these operations satisfied the increase in demand for water in Riyadh City in the 1960s. By the 1970s, even more water was required. In 1978, the Salbouk field was developed to supply Riyadh City with water from the Minjur aquifer. Development of the Buaib field, also tapping the Minjur aquifer, followed in 1979. As the water level fell and demand increased, the Wasia aquifer had also to be tapped. In 1982, 62 wells were constructed. These wells, however, could not supply all the water required for both the city and the region. By 1987, seawater processed in desalination plants provided over half of the region’s potable water supply. A similar trend in water consumption occurred in almost all major cities in Saudi Arabia.

The major urban areas in the country are currently engaged in massive desalination of seawater, at a very high cost, to augment inadequate groundwater supplies. In addition, large investments are being made in sewage and wastewater treatment facilities to bolster existing water supplies through recycling. These investments were initially made in a climate during which oil revenues seemed boundless. Government budgets were notional in the sense that they did not reflect a concern about resource availability. Now, however, the era of limitless resources has passed, and policy-makers are being forced to take seriously the constraints on revenues available for water development.

Table 2 Desalinated water production by major plants in Saudi Arabia

Source: SWCC Annual Report, 1991-92.

Desalination plants have been used on a large scale since 1970. In 1970, the government began construction of 22 large-size desalination plants, spending a total of SRls 51 000 million to augment the supply of good quality water for potable use. The plants currently represent the largest capacity for desalination in the world, with a potential of about 700×10⁶m³/year, or some 30% of world productive desalination capacity (Al Riyadh, 9/2/1992). Currently, four additional desalination plants are under construction, that will add another 1×10⁶m³/day to existing capacity to supply additional water for expanding demand. Table 2 shows the total desalination potential of the major plants in Saudi Arabia.

Most of the desalination capacity was and is provided through the use of the reverse osmosis (RO) and the multistage flash distillation (MSF) processes.

In MSF, seawater is pressurized and heated to maximum desalination plant temperature. The heated liquid is then discharged into chambers maintained slightly below the saturation vapour pressure of water, so that a fraction of its water content flashes into steam. The flashed steam is stripped of suspended brine droplets as it passes through a mist eliminator, and then condenses on the exterior surface of heat transfer tubing. The condensed liquid drips into trays as water suitable for human consumption.

RO uses hydraulic pressure as its energy source. In contrast to MSF, which operates in the range of 50°-115°C (125°-240°F), this process operates at ambient temperatures (below 40°C (104°F)). In this process, a fraction of the pure water content of seawater or brackish water is driven under pressure through a semipermeable membrane, generally of organic material.

The seawater used varies in TDS and salinity. TDS values range from 40 900 ppm at the Al-birk plant in the southern part of the Red Sea to 42 210 ppm at Hagal on the Gulf of Arabia, in the northern part of the Red Sea. Salinity in the Gulf varies from 40 000 ppm to 60 000 ppm depending on the location, currents and time of year. In general, in the Gulf - where the Jubail Plant is located that feeds Riyadh with water - the seawater has more salinity than ocean seawater, which usually averages 35 000 ppm.

In Riyadh City, processed seawater is not directly consumed. Rather, desalinated water is first mixed with groundwater, extracted mainly from deep wells. There are 125 of these deep wells and, along with 34 shallow and medium wells, they supply the remainder of the water consumed in the Riyadh region. This groundwater and the water from the desalination plants means that the water consumed by municipalities in the Riyadh region in 1987 represented 55% of the total water consumed in the country by all municipalities.

Table 3 Reclaimed wastewater use

Source: Olyan, 1996.

Urban effluent waste is also treated and used for agricultural, industrial and landscape uses. Table 3 shows the wastewater reclaimed in Riyadh City and its use. Total water treated is 138 000 m³/day. Agriculture and landscape uses use 118 000 m³/day. Besides the environmental benefits from treating urban wastewater, areas whose wells have dried up, such as Daraiah, were helped by this project.

4. WATER DEMAND

Water demand can be divided into three categories based on use, namely agriculture, livestock and municipal. Municipal demand for water can be further categorized by use - as potable or industrial - or by source - groundwater or from desalination - or by user location - region or city. This section presents the demand for water in Saudi Arabia. The desalination programme is also discussed, as well as the pricing system for water within the Kingdom.

According to the planning documents, total water demand in Saudi Arabia increased from 1 750×10⁶m³ in 1975 to more than 8 600×10⁶m³ in 1985 (MOP, 1990: 170) and to more than 18 200×10⁶m³ in 1995 (MOP, 1995: 192). Although these figures represent a high growth rate in water use, they still underestimate the true demand change for water that was taking place. This water was extracted mostly from non-renewable ground resources, coming from Saudi Arabia’s eight principal aquifers and nine secondary aquifers. The distinction between the principal and secondary aquifers is based on their hydrologic properties and areal extent. Principal aquifers have greater permeability and larger yields than secondary aquifers. However, recharge is small in all aquifers, estimated to be 1 270×10⁶m³/year. Fortunately, the initial amount of water storage is large, totalling about 340×10⁹m³ in primary aquifers and 160×10⁹m³ in deeper, secondary aquifers (MOP, 1985).

4.1. AGRICULTURAL WATER DEMAND

The demand for water comes overwhelmingly from irrigation for crops, principally wheat and fodder crops. At current rates, more than 92% of the total water consumed is used in agriculture; urban sector use is about 8%. The amount of water used for irrigating crops is inextricably linked to agricultural policies and the style of production in the farming sector.

Table 4 shows the water demand in the farming sector, primarily the arable sector, as water used by livestock amounts to less than 5% of all water used in agriculture. Wheat is the predominant user of water in agriculture on an absolute basis, with every hectare on average consuming 13 173 m³ of water and total use accounting for about 48% of the all water consumed in the agricultural sector. The second largest is fodder crops, at 31% total agricultural water consumption. This use shows the importance of the government’s wheat policy as a relatively high-water-consumption programme in relation to water consumption in the country as a whole.

Table 4 Agricultural water demand in 1991

Source: Areas are from MFNE, 1992.

Not surprisingly, the groundwater levels in most regions have fallen by 8 to 15 m during the past ten years. As a result, deeper wells, more powerful pumps and increased energy are needed to support agriculture per unit of output (Kalthem, 1978; British Arabian Advisory Company, 1980). Moreover, the salinity of the principal aquifer (Wasia) is expected to rise to 1 500 mg/l, so it will be necessary to desalinate water just to reduce its salinity to the 500-700 mg/l level considered suitable for potable uses (MAW, 1984).

Planning documents from MOP have continually underestimated water requirements in agriculture. For example, the Second Development Plan (MOP, 1975: 105) estimated that the demand for irrigation water would increase from 1 960×10⁶m³ in 1975 to 2 577×10⁶m³ in 1980, reflecting a 32% increase in consumption over a five-year period, or 6.5% every year. A similar estimation of a moderate increase in the demand for irrigation water was expressed in the Third Development Plan (MOP, 1980: 119). It was not until the Fourth Development Plan (MOP, 1985: 139) that the underestimation of irrigation water was acknowledged:

“Agricultural water demand rose fastest during the Third Plan. From less than 2 000×10⁶m³/year estimated in 1980, water consumption in agriculture rose to an estimated 7 480×10⁶m³/year in 1985, or 84% of total consumption. This rate of water consumption in agriculture is almost four times greater than anticipated in the Third Plan... over 70% of this water comes from the Kingdom’s non-renewable groundwater resources.”

In the Fifth Development Plan (MOP, 1990: 170), MOP estimated that the demand for irrigation water would be at 14 580×10⁶m³ in 1990, almost double the amount of 1985, while in the more recent Sixth Development Plan (MOP, 1995: 193), MOP estimated that demand for irrigation water would be at 16 400×10⁶m³ in 1995. This estimate still appears low. As shown in Table 4, the total water demand by the agricultural sector - based on 1991 areas - is 23 896×10⁶m³, a figure higher by about 45% than the figure forecast. Furthermore, if irrigation averages 16 792 m³/ha, 1.5×10⁶ ha of cultivated land would be expected to use about 25 188×10⁶m³/year of irrigation water, a figure 52% higher than the estimate of the Sixth Development Plan. It shows that the development plans have systematically used lower water consumption rates and in some cases underestimated areas.

Such increases in water demand from agriculture could have been expected by simply comparing the number of wells in the country over time. Table 5 shows the number of private wells by region. Private wells for agricultural purposes in general in the country grew in number by 35% between 1985 and 1989.

Table 5 Private wells by region (1985-89)

Source: MFNE, 1990.

As mentioned earlier, demand data for the years 1980-90 underestimated the actual water consumption in both urban and agriculture sectors, thus not accounting for the real withdrawal rate during that period. For 1995, the agricultural water demand was estimated at 16 400×10⁶m³, a figure much lower than that of 23 896×10⁶m³ for 1991 areas in Table 4. Nevertheless, water demand in fact decreased during 1995. This decrease was a result of intended and unintended policies. In 1994, wheat prices were reduced by the government, from SRls 2 000/t to SRls 1 500/t. However, more importantly, the price of the principal input for pumping water - diesel fuel - was tripled, thus making pumping water far more expensive than before. Furthermore, due to a budget deficit and capital shortage, the government has been forced to delay payments to wheat producers for four years on average since 1991, thus in effect discounting announced prices at four-year periods.

These decisions created an unclear atmosphere about agricultural policy. Marginal wheat producers exited the market due to the increased cost of pumping water and the decreased returns from producing wheat. Other producers were willing to produce at the new low profit margin, but only if payment were made within a year of wheat delivery. As a result, wheat output declined from its peak of 3 800 000 t in 1992 to about 1 500 000 t in 1995, which helped to reduce demand for water. However, water demand in agriculture is still higher than its average in the 1980s and higher than any sustainable rate.

4.2. MUNICIPAL WATER CONSUMPTION

Whatever its category, water demand for municipal use in Saudi Arabia has increased rapidly during the past fifteen years. According to the Fifth Development Plan, water demand for municipal and industrial uses increased over threefold, from 510×10⁶m³ in 1980 to 1 650×10⁶m³ in 1990 (MOP, 1990). Similarly, the Sixth Development Plan indicates that water demand for municipal and industrial uses increased to 1 800×10⁶m³ in 1995 and that demand will increase to 2 800×10⁶m³ in 2000 (MOP, 1995: 193). This increase was prompted by increasing urbanization and industrialization, along with Saudi Arabia’s high annual population growth of 3.6% over the past ten years, and has meant that water requirements in the non-agricultural sector have more than tripled. Natural aquifers do not supply all the water necessary for the growing Saudi population; instead, desalination plants are used to satisfy this increase in water demand.

By 1987, 54% of the 625.5×10⁶m³ of total water demand by municipalities was supplied by desalination plants, of which 62% went to the Riyadh region (Table 6). Water demand for Riyadh City alone during the summer of 1996 reached 1.35×10⁶m³ daily; for a population of about 3 million, that amounts to about 450 l/caput/day. Considering this rapid increase in water demand, the expected population of Riyadh City of 4.2 million in 2010, and the limited capacity of desalinated water pipelines from the Gulf, the government will have to review its water plan. In particular, the high cost of providing potable water could stimulate a review of current policy.

Table 6 Origins of municipal water supplies in Saudi Arabia in 1988

Source: MOMRA, Municipalities Statistics, 1988.

Other major cities are increasingly dependent on desalinated water, as groundwater levels fall due to the increasing use of such water in agricultural activities. In response, the government is extending water pipelines from desalination plants to these cities to carry desalinated water. Other towns and villages in traditionally agricultural areas are also becoming more heavily dependent on deep water resources due to the dropping levels in shallow and medium wells, and deteriorating quality, and becoming increasingly candidates for desalinated water supplies.

Table 7 Comparative estimates of desalinated water costs

Note: (1) MSF = multistage flash [distillation]; RO = reverse osmosis
Source: IDA, 1991.

Table 8 Comparative water rate structures in various countries

Sources: The Economist, 1991; Adil Bushnak, 1992.

The costs of desalinated water in Saudi Arabia are relatively high, averaging $US 1.01/m³ (IDA, 1991), as illustrated in Table 7. Such a cost figure is consistent with other data that show that in general the cost of desalinating a cubic metre of seawater in Saudi Arabia ranges from $US 1.01 to $US 2.50. Internationally, various studies have been done to survey existing desalination plants and technologies, and to estimate their costs. A comprehensive reference in the field is IDA (1991). The Saudi Arabian estimates, however, do not mean that the price of water in Saudi Arabia reflects the costs of desalination, or even just simple extraction. Rather, water in Saudi Arabia has been supplied free of charge to industrial and agricultural users, while charges to domestic users have been heavily subsidized (MOP, 1985: 143). The 22 desalination plants in the country, which cost SRls 51 thousand million to build ($US 13 600 million), have a production capacity of 693×10⁶m³/year. Desalinated water costs SRls 4.89/m³, yet is sold at a block increasing rate of SRls 0.25 to 2.0/m³; at such prices, municipalities consume about 345×10⁶m³ - 62% from desalination and the rest from groundwater resources.

Table 8 shows the prices of domestic water that the government charges and compares these prices with those current in other members of the Gulf Cooperation Council and in other countries. For potable water, prices in Saudi Arabia appear to be the lowest of all countries, especially since the first bracket of users - i.e., the 0-100 m³ range - is large and constitutes about 92% of all water users. The price of potable water in Saudi Arabia is only 8% of its price in the USA, 2% of its price in Australia, 4% of its price in the United Arab Emirates, and 3.6% of desalination cost. This single statistic compared with the cost of desalination shows that there is a high welfare cost implicit in the current rate structure, whereby industrial users are not charged and low rates are charged to 92% of users.

These large subsidies encourage waste of a costly commodity. Hence, water conservation is a key concern of policy-makers. A good means to conserve water would be to implement a policy based on market incentives. Little effort has been made in this direction. In 1990, a block-increasing price scheme was developed to encourage more rational use of expensive desalinated water in municipalities, but it is based on initial high subsidized rates and covers only 3% of water use. Hence, as a water conservation measure, it has been ineffective.

Table 9 Summary water balance (×10⁹m³/year)

Sources: MOP, 1990: 179; MOP, 1995: 193.

In sum, a large gap exists in Saudi Arabia between use and recharge rates. Evidence of dwindling water stocks includes the drying up of natural springs in some regions and the ongoing fall in water levels in all types of aquifers. As of 1981, groundwater stocks were estimated to be about 370 000×10⁶m³, with an annual recharge of 319×10⁶m³. Water consumption by municipalities is 345×10⁶m³/year, of which 62% is supplied by desalinated water at a cost of between $US 1.01 and 2.50/m³ (SRls 3.79 to 9.375/m³), but sold by municipalities at a block increasing price that ranges from SRls 0.15 to 2.00/m³ for most users, or at about 4 to 20% of its desalination cost. Much more important for total water use, 6 900×10⁶m³/year - 95% of the total groundwater consumption - is for agriculture use.

Table 9 presents a summary of the national water balance in Saudi Arabia based on figures from the Fifth and Sixth Development Plans (MOP, 1990: 170; MOP, 1995: 193). As mentioned earlier, demand data for the years 1980-90 underestimate the actual water consumption in both urban and agriculture sectors and so do not accurately reflect the real withdrawal rate during that period. However, it may be assumed that the water consumption rate for 1995 is valid and that it will continue for the coming 15 years; in other words, conservation efforts will offset growth rate by population increase. From the above, the 14.84×10⁹m³ that is currently extracted from non-renewable resources is going to exhaust the principal remaining water reserves listed in Table 1 in about 13 years, and the total remaining reserves in about 19 years.

5. INSTITUTIONS AND POLICY OBJECTIVES

5.1. POLICY OBJECTIVES

The Government of Saudi Arabia, in its planning documents, has stressed three major objectives: economic diversification; income distribution; and food security. This section discusses each objective briefly, and observes that they have often been mutually inconsistent.

Like many development plans, governmental objectives stated in various planning documents have often been contradictory. For example, efforts to create a more diversified economy have been hampered by a commitment to free trade and free markets, resulting in specialization in oil exports. In addition, the objective of diversification through agriculture has been made difficult by the spread of consumption subsidies.

This tension among objectives is apparent in the national food strategy, especially policies implemented at different times. However, policies implemented were consistent with the Second Development Plan (MOP, 1975: 5), which stated that it was government policy to

“make essential goods, especially food items, available at stable and reasonable prices, subsidizing prices if necessary, with due regard to the effect on domestic production”

In 1973, the government began subsidizing food imports of basic commodities equivalent to the difference between the world price in 1973 and current prices. This policy insulated the internal market from any instability in the world market. On the producers’ side, the government response came rather late, in 1978, with a package of generous producer subsidies. Even that did not have the expected effect on the shrinking agriculture sector, until 1980, when more fallow land was distributed, financial facilities were made more available, and much of the investment in non-tradables (basic infrastructure) was in its final stages. The result of all of these policies cannot be called fully planned. Nor can the interactions have been fully interpreted. In brief, Saudi Arabian agricultural policy is characterized by subsidies: subsidies to outputs, subsidies to inputs, and subsidies to natural resources.

5.2. POLICY INSTRUMENTS

The government, in its efforts to achieve its objectives, has used a number of policy instruments. These instruments affect output prices as well as input prices and water use.

5.2.1. Policies affecting output prices

Output prices of some crops are set and guaranteed by the government. For example, the domestic wheat price was set at SRls 3 500/t, then reduced to SRls 2 000/t, and then finally reduced to SRls 1 500/t, and the Grain Silos and Flour Mills Organization (GSFMO) buys all the stocks. Barley, priced at SRls 1 000/t, is also purchased by GSFMO. All supplies are purchased by GSFMO as well. In addition, a production subsidy is paid by MAW to producers of some agricultural goods. For example, date producers are given a subsidy of SRls 0.25/kg, even though they sell their output in the market. Date producers are given an additional subsidy of SRls 50 per tree planted. The effect of these incentives on production is evident from the growth the sector has experienced in the past ten years.

5.2.2. Policies affecting input prices

Other types of policy instruments the government is using to influence production are subsidies on fertilizer, machinery and equipment, and irrigation systems, as well as subsidized credit. Fertilizer is subsidized by MAW by 50% of its cost to the farmer. Seeds of most crops are subsidized at 50% of their cost, and in some cases high-yielding varieties are imported by the government or specialized companies for distribution to farmers. Fuel is subsidized in that the domestic price of diesel is set lower than the CIF price at the port. All agricultural machinery (e.g., tractors, ploughing and levelling accessories, harvesters, balers and combines) is subsidized 45% by the SAAB. Irrigation systems (e.g., pumps, engines) are subsidized 45% by SAAB.

Livestock production is also subsidized through various policies. Barley and other concentrated feed components are sold to livestock growers at below market price. A subsidy is paid to livestock raisers at a rate of SRls 30 per head of lamb and SRls 80 per head of camel. The air transportation costs for imported dairy cows are paid by the government. A subsidy of 30% is given on the cost of all poultry, eggs and milking equipment.

Farmers also are provided with free credit facilities for short-, medium- and long-term needs. Short-term, one-year loans for purchases of seed, fertilizer, fuel and feed, and for machinery rental, are available from the government. Medium-term loans, which must be paid back within ten years, cover machinery, equipment, storage, building, well drills, well deepening and cleaning, plants, and other fixed capital investment costs. Long-term loans are usually paid back over 10 to 25 years. Such loans are aimed at development of fallow land, including the associated costs of well development for large-scale projects.

In sum, these incentives affecting output and input prices have a direct impact on the use of natural resources, be they land or groundwater resources. Direct incentives affecting the use of natural resources include free land and water. Land is distributed free by the government to farmers. Water is free and comes from the country’s groundwater resources. Other investment subsidies have an indirect effect on resource extraction rates. For example, irrigation system subsidies, and lowering the cost of pumping increase the profitability of pumping. The output subsidies especially encourage water-intensive crops. Individually and collectively, government policies encourage the use of fossil groundwater.

5.3. PROPERTY RIGHTS

5.3.1. Land rights

The current land ownership patterns in Saudi Arabia have been influenced by history, customs and religion. Distribution of land for agriculture and settlement in modern Saudi Arabia was started by King Abdulaziz as early as 1912, during the period of unification of the country, which was between 1899-1931 (Hajarah, 1982; Beaumont and McLachlan, 1985). In 1968, the government issued the Public Land Distribution Ordinance (PLDO) (Royal Decree No. 26/2, 1968), which ruled that undeveloped lands are owned by the government and appropriation of such lands by other parties is not recognized. The PLDO gave MAW the power to organize the distribution of public land. Individuals were to be allotted 5 to 20 ha, while companies and individuals with agribusiness projects could receive up to 400 ha.

In accordance with the fallow land distribution law, 1 519 928 ha were distributed from 1968 to 1990, although 90% of the distributions occurred during the 1980s. Some 56.6% of the total land distributed, or 859 306 ha, was employed in agribusiness projects.

Individuals who benefited from the distribution of land numbered 67 686, and held 399 244 ha, or 26.3% of the distributed land, while 17 agricultural companies accounted for 261 378 ha, 17.2% of the distributed land (SAMA, 1991).

5.3.2. Water rights

Control over natural resources in Saudi Arabia is defined by Islamic Law. A fundamental characteristic of ownership in Islam is the concept of supreme ownership. Land and water, in general, are owned by God, and his servants (mankind) are to share its utilization based on their capabilities and needs. In particular, this ownership of land is limited to lands that have been cultivated before. For water, however, the case is different. Specifically, all water is considered God’s property, and its ownership is possible only when effective possession of water takes place, which is the enclosure of water in a container such as a jar, a tank, a pool or any receptacle. All other forms of possession are merely the right to use without actual ownership. Only an owner, however, can sell water. This water must be in receptacles (FAO, 1978). Priority of water use is first for the people, then livestock, then agriculture and industry, and finally other uses (MOP, 1990: 63; Al-Rasheed, 1983).

Throughout history, a general accepted principle has been that the owner of land has absolute ownership and unlimited right to withdraw any water that is beneath his land. This understanding is called “English Common Law,” and it is still prevalent in most countries. In general, Islamic principles also recognize the ownership of groundwater by a landlord, but limit the owner to reasonable uses at certain farms and provide special protection for earlier users against subsequent pumpers and for all pumpers against those who contaminate the aquifer’s water (FAO, 1978; Al-Rasheed, 1983). This system is similar to the North American or “reasonable use” doctrine.

In a recently announced constitution, otherwise known as the Governing System of Saudi Arabia, Article 14 stipulates that all resources underground are owned by the government (NY Times, 2 March 1992; Tijart AlRiyadh, 31(354): 12). The article does not explicitly state whether those resources include groundwater, but it could be inferred that as long as effective possession of water does not occur, then the water is considered public property. This is consistent with Islamic law.

These theoretical and practical differences in rights to natural resources depend on who is making use of the resources where, theoretically, priority of water use is first for the people, then livestock, then agriculture and industry, and finally other uses. Practically, agriculture currently controls the resource, while municipalities augment their water supplies using desalination.

6. OPTIMAL ALLOCATION

The eventual exhaustibility of groundwater resources is evident from the falling water table of major agricultural areas due to agricultural, industrial and urban consumption. This fall, accompanied by increasing extraction costs, means that depletion affects the cost of future activities. Given the finite nature of the groundwater resource and the fact that the cost of extraction is correlated with the cumulative volume of past extraction, this is equivalent to saying that the cost of groundwater extraction is negatively correlated with the volume of the remaining stock of groundwater. This concept is commonly referred to as the “stock effect” in natural resources literature. Such terminology emphasizes the fact that any increase in cost is a function of remaining stock of the resource rather than the amount of the resource extracted in any one time period. The ceteris paribus assumption usually used in static analysis is no longer appropriate because variables in the past, present and future are all linked. Hence, the assumption of no change in current values is inconsistent with the nature of the resource dynamics.

To guard against water shortages, the government has resorted to massive desalination of seawater to augment the inadequate supply to urban areas from groundwater. Not surprisingly, there are serious concerns that distorted agricultural policy prices have a negative effect on the socially optimal rate of groundwater extraction. A more efficient set of policies would be to reduce the demand for water in agriculture, perhaps opening the way for the transfer of water from the agricultural sector to urban use countrywide. Such transfers of water from agricultural activities to non-agricultural activities are likely to have a positive economic welfare effect and ensure better allocation of these limited resources. Optimal water allocation has four dimensions:

(1) Spatial allocation: the efficient allocation of water between competing uses in the same sector but in different geographic areas.

(2) Sectoral allocation: the efficient allocation of water between uses based on purpose, namely agricultural or urban (municipal and industrial).

(3) Intertemporal allocation: the efficient allocation of water over time.

(4) Conjunctive allocation of water: the efficient allocation of water based on its source, whether from the surface, from the ground, or from backstop technology (desalination or treated urban effluent).

Were policy-makers utility-maximizing, forward-looking planners, they would try to allocate available resources between current and future uses to ensure that marginal revenues from water use are equal across all four categories: space, sector, time and source. Under a well-defined system of property rights and reasonably functioning product and resource markets, government intervention would not be required to achieve optimal use. Users earning low marginal net benefits would trade their rights to those who would earn higher net benefits from the water, benefiting both individual users and society as a whole.

Figure 2 Effect of policies on water benefits and costs

In such a well functioning market for producing and consuming water, there will be no divergence between marginal private cost (MPC) and marginal social cost (MSC) for producing or pumping water. In other words, MPC = MSC. Similarly, there will be no divergence between marginal private benefits (MPB) and marginal social benefits (MSB) from consuming water, so MPB = MSB. This would be true across use sectors (agriculture and urban) and across use regions, taking into account shipment cost. Furthermore, the current MSC value of producing 1 m³ of water is equal across time periods of the planning horizon, and the current MSB value would be equal across time periods.

Figure 2 illustrates the effect of the government policy on water use in Saudi Arabia. Government subsidies to output prices shifts MSB from using water to the right hand side relative to MPB. Thus, making water private added value higher than water producing activities. Input subsidies, in contrast, reduce the cost of pumping water, hence shifting MSC to the right relative to MPC, thus making pumping more water cheaper than before. The water quantity Q_s represents water optimal static extraction when prices and markets are efficient, while water quantity Q_p represents water static extraction with the distorting subsidies. The difference between Q_p and Q_s, namely (Q_s - Q_p) represents the policy impact on water extraction.

In Saudi Arabia, none of the conditions required for optimal use is present. MPB is seriously distorted as a result of distorted agricultural policies. For example, both produce and inputs markets are seriously distorted by government subsidies. The absence of water markets or the regulation of groundwater also complicates the allocation problem. Unlike surface water, the movement of groundwater is virtually impossible to control. It responds to pumping by moving in the direction of the pumping node and recharges by moving away from source of the recharge. Because property rights cannot be assigned to this type of fugitive resource, farmers have no incentive to restrain their pumping. The perceived marginal cost of pumping underestimates the real increase in costs as all producers react in a similar fashion. The result is an overexploitation of the resource. No serious attempt to control water withdrawals has been made, although the government now has a pump-licensing scheme.

Figure 3 Water reserves and production pattern

The current policy of using most of the groundwater resources to produce subsidized agricultural products, while producing expensive desalinated seawater for urban areas, can be presented graphically in a way to illustrate the static welfare loss involved with such policy. In Figure 3, government input subsidies shift MSC of pumping from MSC to MPC of extraction, thus making pumping each unit of water much cheaper than its social cost. Similarly, government output subsidies shift the MSB to the MPB, thus making pumping water profitable. In fact, under MSB the principal aquifers will not be tapped. Furthermore, the extraction rate will be much lower than under the MPB curve. The marginal cost of desalination of seawater (MC), is an upper bound on all the groundwater marginal cost of pumping.

The derived optimality conditions state that for each resource in each district the marginal benefit of a unit of groundwater must equal its marginal cost of pumping plus its user cost. The user cost, or the cost of depleting a resource stock, consists of the value of foregone income at the terminal time due to resource depletion. The marginal benefit of desalinated seawater must equal its marginal cost.

There is a substantial literature dealing with natural resource economics in general and groundwater in particular. A comprehensive account up to the late 1970s is found in Dasgupta and Heal (1979). It is particularly applicable to Saudi Arabia because the difference between the marginal value of 1 m³ of water retained in the ground (the remaining stock) as a substitute for 1 m³ of flow water pumped in the present period (similar to the extraction rate) is critical to water allocation in the long term. Research focusing on groundwater issues can be found in Burt (1964, 1966, 1967); Burt and Cummings (1970); Gisser (1983); Feinerman and Knapp (1983); and Lemoine (1984), although they do not deal directly with the role of backstop technology.

The problem of intertemporal allocation was analysed by Renshaw (1963). He concluded that the annual saving per unit of water retained in the ground depends directly on three factors:

(1) the long-term marginal cost of extraction of one unit of water per one unit of elevation;

(2) the amount of water to be extracted in future periods; and

(3) the amount of water saved for future extractions as a result of leaving one unit of the resource in the ground.

Renshaw added that if no monetary value is attached by farmers to water in the future, i.e., if no property rights are defined, farmers will not have an incentive to maximize the present value of groundwater over time and would behave rationally in maximizing their short-term benefit functions.

Consider an example of a constant demand for water. The efficient extraction path involves declining use of groundwater over time because pumping would have to stop when the water table ran dry or the marginal extraction cost (the cost of pumping water to the surface) was either greater than the marginal benefit of the water pumped or greater than the marginal cost of acquiring water from a substitute source. Effectively, the ultimate substitute would be then desalination, the backstop technology. Its cost would be an upper bound on the marginal cost of extraction.

6.1. BIAS TOWARD INTENSIVE USE OF WATER

A large gap exists between extraction efforts and recharge rates. The water level has been declining for years, natural springs in some regions have dried up, and many wells have gone dry. Output price subsidies include those for the two of the most important crops, namely, wheat and barley. These subsidies shift the demand functions upward for these crops to levels substantially above that of their border price. Subsidies on agricultural inputs include those on fertilizer, machinery and equipment. These subsidies shift the cost functions in agricultural activities substantially below those of border prices.

Subsidies on irrigation systems include those on pumps and engines at 45% of their cost. Fuel needed for the operation of water pumps is sold at subsidized prices. Interest-free loans can cover up to 80% of the capital requirement for pumping systems. These subsidies to irrigators are best understood with the help of farming system models.

In summary, the value of the water as an exhaustible resource needs to be accounted for in government policy. Significant opportunities exist for adjusting government agricultural policy to reduce excessive groundwater extraction. Desalinated water should not be used to satisfy municipal water demand; rather, groundwater should be used instead. Water exploitation is excessive, primarily because of the distortion from government agricultural policies.

6.2. POSSIBLE POLICY CHANGES

6.2.1. Reduction in wheat subsidies

Among the agricultural price policies leading to an overexploitation of water, the most important policy intervention concerns wheat. The wheat price policy is important for policy-makers not only because of the priority it places on food security and income distribution, but also because of the expense of the programme. In 1991, production of wheat in Saudi Arabia reached 3 800 000 t. With domestic consumption at only 1 100 000 t, and with the nearest export market, the other Gulf countries, able to absorb only about 400 000 t, Saudi Arabian production was well in excess of Gulf countries’ needs. The direct budget cost of Saudi Arabian wheat production was SRls 7 000 million, yet, had the wheat to satisfy local demand been purchased on the world market, the expenditure would have been only SRls 800 million.

The huge subsidy, about 5% of the national budget, is usually justified on the grounds of self-sufficiency. Another justification is income distribution, since the money is directed toward the highly populated rural economy (Al Riyadh, 1992). In reality, however, only some 25 000 farmers participate in the wheat programme countrywide. Furthermore, about 50% of wheat output comes from large farmers, those who produce more than 200 t/year. Large farmers constitute only 10% of all the farmers participating in the programme and only about 1% of all farmers in Saudi Arabia. In other words, 50% of the value of the programme is received by 1% of all the farmers. Moreover, the large farms are owned and managed by urban residents; the work is performed by foreign labourers and expatriates. In short, the wheat subsidies have not had the distributive effects desired by government’s planners.

Table 10 shows the number of farmers participating in the wheat programme in 1988 and 1989, their output and its value.

Table 10 Wheat producers and their production in 1988 and 1989

Source: GSFMO Annual Report for 1988-89.

The situation of unequal distribution of benefits from the wheat programme regionally and by size of farms means that any rationalization of the present policy must cover both policy management and incentives, assuming that the government maintains some non-efficiency objectives. If the government decides to reduce production of wheat to 1 100 000 t to simply meet the country’s domestic demand, this decision would satisfy the political objective of self-sufficiency in wheat production.

If the government is also interested in decreasing the use of irrigation water, the reduction in wheat delivery must be accompanied by reduction in total irrigable land, as a set-aside. This action would not only help conserve water but also preserve soil quality by encouraging proper crop rotation. Hence, this policy would be greatly desirable from the point of view of the non-efficiency objective of self-sufficiency, since it would increase the time horizon of extraction of existing groundwater resources and consequently prolong Saudi Arabian self-sufficiency in water.

Scenario 1 introduces a reduction in wheat price to SRls 1 000/t and reduces area and the irrigable land set-aside both to 50%. Wheat output decreases more than before. Specifically, wheat production is only about 1 900 000 t for the whole country. The budget cost under this scheme dramatically decreases from SRls 6 700 million to SRls 1 890 million, a tremendous savings of 72%. The amount of wheat produced, however, would satisfy all domestic demand as well as that of the Gulf states. In fact, surplus wheat amounting to 300 000 t would even cover any increase in demand or could be used as donations to other countries. The storage capacity of the government purchasing agency - 2 850 000 t - would be underutilized, but could serve as a strategic stockpile storage facility.

Scenario 2 is the most ambitious and efficient policy. Here, the government discontinues its wheat programme, which means that Scenario 2 is the laissez faire option. Hence, all wheat would be imported from the world market. These imports would total 1 100 000 t/year and would cost SRls 495 million, thus saving the government both SRls 6 600 million and 11.6×10⁹m³ of water. If the budgetary savings were used for forgiving farmers loans from SAAB, all loans would be paid off in two years. Further welfare gains could be realized, if water saved is diverted to urban areas to substitute for the more costly desalinated water.

6.2.2. Privatization of municipal water supplies

The current water supply situation is such that the government is desalinating seawater to provide water for municipal use in most cities. For example, desalinated seawater is transported via 470 km of pipeline and six pumping stations from the Gulf Coast to Riyadh City. The pumping stations are needed because Riyadh City is 620 m above sea level. This entire operation is accomplished at a very high cost. For example, the Jubayl II Plant cost more than SRls 6 400 million ($US 1 700 million) (Al-Ahd, in SWCC, 1992) and will produce water at a cost of at least SRls 4.2/m³ ($US 1.10/m³) at the point of production (Leitner, in IDA, 1991). An additional SRls 0.75/m³ ($US 0.20/m³) is required for transportation to Riyadh City. Yet desalinated water is not the only source of potable water available in the Riyadh region for municipal use. Rather, groundwater is available in areas within 60 km of Riyadh City.

Currently, the government has a monopoly over the supply and production of groundwater and desalinated water to municipalities. Privatization of water supply to municipal authorities could be more cost effective. In particular, allowing companies to compete in establishing desalination plants and supplying water at a per-unit-of-output rate to the municipal authorities would probably lead to cost reduction, technical innovation and development of alternative sources of water supply. Moreover, allowing farmers in close proximity to urban centres to sell water to municipal authorities could prove to be economically beneficial to both. In this regard, several questions are of interest to policy-makers:

(1) At what price/m³ would farmers start selling water to the municipal authority?

(2) If the government paid farmers what it costs to desalinate one cubic metre of water, what would be the effect on the production plan, income and water use on farms?

(3) What are the trade-offs between each policy?

The value of a unit of water produced should be set at SRls 4.89/m³ ($US 1.30/m³), the cost to produce and transport desalinated seawater to Riyadh City. Farmers, however, are not allowed to increase their water consumption (or production) per hectare above that of wheat producers because the government does not want to use more water than at present. So, the water yield of every hectare devoted to the water-selling activity cannot be greater than 13 170 m³/ha/year, which is how much a hectare of wheat uses during its production season. The costs of pumps and diesel engines as well as all operating costs are subtracted from revenues. In addition, the cost of water purification and delivery to the farm gate - SRls 1 875 for 13 170 m³ - is incorporated in this exercise.

Farmers in this case would prefer to sell water rather than plant their land as long as the land set-aside condition is imposed. Small farmers would sell an average of 492 000 m³/year at a price of SRls 4.9/m³, while large farmers could sell much more. Even though the Saudi Arabian government paid SRls 4.89 for the water, it would save on average SRls 166 542/year from foregone wheat production per small farm and SRls 7 021 890 per year from foregone wheat production per large farm. The government could reduce the price of water to SRls 1 000/m³ for large farmers and to SRls 850/m³ for small farmers, and it still would be profitable for farmers to produce water instead of wheat, for sale to municipal authorities. At SRls 1 000/m³, the government would save in two ways: first, the cost of water would decrease by SRls 3.9/m³, and, second, the cost of the wheat programme would decrease by about SRls 9 000 for each hectare set-aside, a total of about SRls 5 500 million/year for the whole country.

6.2.3. Ministry of Water

Currently, there is no ministry with sole responsibility for water affairs in Saudi Arabia. Water services and issues are handled by three different agencies, namely MAW, the Ministry of Municipalities and Rural Affairs (MOMRA) and the Saline Water Conversion Corporation (SWCC). MAW handles most of the work and policies related to water. MAW is also responsible for completing the National Water Plan, which was delayed because of insufficient reliable information about water supplies. At the same time, it is responsible for promoting the interest of farmers and the agricultural sector as a whole. MAW is the agency that issued the drilling permits that tapped most of the shallow and deep aquifers. The Water Affairs Department within MAW is the department responsible for well digging, construction of dams, studies, hydrological surveys and water reuse. In general, MAW is responsible for providing water and building water networks, which are then transferred to MOMRA to administer. MOMRA is responsible for installing home connections, and for wastewater treatment. SWCC is a government agency that builds new desalination plants, and at the same time operates and maintains existing plants.

Organization of the government water sector is not clear and may have areas of overlap. Organizing into a single ministry these agencies currently in different ministries, and providing the consolidated unit with a clear set of objectives about optimal planning of the water sector is likely to be better than leaving the present structure, and in particular now that water issues are increasingly becoming vital to the well-being of the whole economy and the future success of development planning in the country.

In summary, this paper has presented the supply and demand of water in Saudi Arabia and analysed the interaction between agricultural policy and the economics of water use in Saudi Arabia. More investigation needs to be made into the countrywide water dynamics and movement of the resource as well as regional data. Furthermore, aquifers that extend across regional and national boundaries need to be examined. The effect of different extraction rates on the direction and extent of water movement is important to common property resources, especially for transnational aquifers. Water exploitation is excessive in Saudi Arabia. Much of the water extraction in agriculture is driven by policy-induced distortions that affect agricultural inputs and outputs and water pumping costs. Agriculture is the largest water user in Saudi Arabia in terms of volume, but it is a low value, low efficiency and highly subsidized water user. Huge potential welfare gains are lost by delaying required changes in the agricultural policy for even a single year.

BIBLIOGRAPHY

Al-Rasheed, Abdullah. 1983. [The right of using water in Islamic shariah.] (in Arabic) Agriculture Magazine, 14(1, 2, 3)

Al-Saad, Faissal. [No Date]. [Water Irrigation and Crops Production in Saudi Arabia] (in Arabic). Riyadh: Saudi Arabian Printing Co.

Al-Sheikh, Hamad M.H. 1994. Agricultural Policy and the Economics of Water Use in the Riyadh Region of Saudi Arabia. PhD Dissertation. Stanford University, Stanford CA, USA.

Authman, M.N. 1983. Water and Development Process in Saudi Arabia. Jeddah, Saudi Arabia: Tihama Press.

Beaumont, P. 1977. Water and development in Saudi Arabia. Geographic Journal, 143.

Beaumont, P., & McLachlan, K. 1985. Agricultural Development in the Middle East. New York NY: John Wiley.

British Arabian Advisory Company. 1980. Water Resources of Saudi Arabia. Volume 1.

Burt, O.R. 1964. Optimal resource use over time with application to groundwater. Management Science, 2: 81-93.

Burt, O.R. 1966. Economic control of groundwater reserves. J. Farm Economics, 48: 632-647.

Burt, O.R. 1967. Temporal allocation of groundwater. Water Resource Research, 3: 45-56.

Burt, O.R., & Cummings, R.G. 1970. Production and investment in natural resource industries. American Economic Review, 60: 576-590.

Bushnak, Adil. 1992. Water privatization. Desalination and Water Reuse, 212.

Dasgupta, P., & Heal, G. 1979. Economic Theory of Exhaustible Resources. Cambridge, UK: Cambridge University Press.

El Khatib, Abdel Basset. 1982. Seven Green Spikes: Water and Agricultural Development. Second edition. Riyadh: MAW.

Feinerman, E., & Knapp, K. 1983. Benefits from groundwater management: magnitude, sensitivity and distribution. American J. Agricultural Economics, 65: 701709.

FAO [Food and Agriculture Organization of the United Nations]. 1978. Water Law in Muslim Countries. FAO Irrigation and Drainage Paper, 20/2.

Gisser, M. 1983. Groundwater: focusing on the real issue. J. Political Economy, 9: 1001-1027.

GSFMO [Grain Silos and Flour Mills Organization]. 1992. Annual Report - 1991-92. Riyadh: GSFMO.

Hajrah, H.H. 1982. Public Land Distribution in Saudi Arabia. London: Longman Press.

Heal, G. 1976. The relationship between price and extraction cost for a resource with a backstop technology. Bell J. Economics, 7: 371-378.

Hotelling, H. 1931. The economics of exhaustible resources. J. Political Economy, 39: 137-175

IDA [International Desalination Association]. 1991. Proceedings of the World Conference on Desalination and Water Reuse. 2 vols.

IFPRI [International Food Policy Research Institute]. 1990. World food trends and future food security. Food Policy Report.

Kalthem, M. 1978. Evaluation of Riyadh City Water Supply and Demand. PhD Dissertation, University of Arizona, USA.

Lemoine, P.H. 1984. Water Resources Management in the Salines Valley: Integration of Economic and Hydrology in a Closed Control Model. PhD Dissertation. Stanford University, Stanford CA, USA.

MAW [Ministry of Agriculture and Water]. Various dates. Various issues of the Agriculture Statistical Year Book. Riyadh: MAW.

MAW [Ministry of Agriculture and Water]. 1982. Census of agriculture according to farm size, 1981-82. Riyadh: MAW.

MAW [Ministry of Agriculture and Water]. 1984. Water Atlas of Saudi Arabia. Riyadh: MAW.

MAW [Ministry of Agriculture and Water]. 1986. Water requirements of important crops in Saudi Arabia. Riyadh: MAW.

MFNE [Ministry of Finance and National Economy]. Various dates. Saudi Arabia Statistical Yearbook. Riyadh: MFNE.

MOP [Ministry of Planning]. 1970. First Development Plan: 1970-1975. Riyadh: MOP.

MOP [Ministry of Planning]. 1975. Second Development Plan: 1975-1980. Riyadh: MOP.

MOP [Ministry of Planning]. 1980. Third Development Plan: 1980-1985. Riyadh: MOP.

MOP [Ministry of Planning]. 1985. Fourth Development Plan: 1985-1990. Riyadh: MOP.

MOP [Ministry of Planning]. 1990. Fifth Development Plan: 1990-1995. Riyadh: MOP.

MOP [Ministry of Planning]. 1990. Seminar on Development and Uses of Water Resources, 1982. 2 vols. Riyadh: MOP.

MOP [Ministry of Planning]. 1995. Sixth Development Plan: 1995-2000. Riyadh: MOP.

MOMRA [Ministry of Municipalities and Rural Affairs]. 1989. Annual Report. Riyadh: MOMRA.

Oren, S., & Powell, S. 1985. Optimal supply of a depletable resource with backstop technology, Heal’s Theorem revisited. Operations Research, 32.

Olyan, Amal H. 1996. [Arab Water Security: An Economic Objective or Political?] (in Arabic) Riyadh: Dar al Ulom Press.

Renshaw, E. 1963. The management of ground water reservoirs. J. Farm Economics, 45: 285-295.

SAMA [Saudi Arabian Monetary Agency]. 1991. Annual Report. Riyadh: SAMA.

SWCC [Saline Water Conversion Corporation]. 1992. Annual Report - 1991-92. Riyadh: SWCC.

Sogreah. 1968. Water and Agricultural Development Studies, Area V. Sogreah, Riyadh.

Sogreah. 1969. Final Report, Parts 1 & 2: Finding and Recommendations, Area V. Sogreah, Riyadh.