Algeria - Agriculture : a new opportunity for growth

Report No. 7419-AL Algeria Agriculture: A New Opportunity for Growth April 5, 1990 Agriculture Operations Division Country Department II Europe, Middle East and North Africa Regional Office FOR OFFICIAL USE ONLY Document of the World Bank This document has a restricted distribution and may be used by recipients only in the performance of their official dut;es. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY AND EQUIVALENT UNITS Currency Unit Algerian Dinar (DA) US$1.00 DA 7.5 (1989 average) DA 1.00 US$0.13 (1989 average) FISCAL YEAR January 1 - December 31 GLOSSARY OF ABBREVIATIONS APF - Accession A la Propri6t6 Fonciere BADR - Banque de l'Agriculture et du D6veloppement Rural CCLS - Cooperatives des Cer6ales et des Legumes Secs DAS - bomaines Agricoles Socialistes EAC/EAI - Exploitations Agricoles Collectives/ Exploitations Agricoles Individuelles EEC - European Economic Community INRAA - Institut National de la Recherche Agronomique d'Alg6rie LSI - Large Scale Irrigation ME - Minist6re de 1'Equipement OER - Official Exchange Rate OAIC - Office Alg6rien Inter-professionnel des C6r6ales O&M - Operation and Maintenance OPI - Offices des P6rimetres d'Irrigation. SER - Shadow Exchange Rate SSI - Small Scale Irrigation UNDP - United Nations Development Program  FOR OFfICLAL USE ONLY ALGERIA AGRICULTURE: A NEW OPPORTUNXTY FOR GROWTH Table of Contents Page Number EXECUTIVE-SUMMARY ..... . . . . . . . . . . . . . . . i PART I. ALGERIAN AGRICULTURE: THE PAST IN PERSPECTIVE AND POTENTIAL FOR GK3WTH I. BACKGROUND ..... . . . . . . . . . . . . . . . 1 A. General Economic Framework . . . . . . . . . . 1 B. Agricultural Overview . . . . . . . . . . . . . 2 II. PAST AGRICUTURAL PERFORMANCE . . . . . . . . . . . . 8 A. Introduction ..... . . . . . . . . . . . . 8 B. Level of Food Self-Sufficiency . . . . . . . . 9 C. Yields ............. ..... . 10 D. The Dual Nature of the Agricultural Sector . .11 III. SOURCES OF PAST GROWTH.14 A. Overview . . . . . . . . . . . . . . .14 B. Changes in Area Cultivated . . . . . . . . . . 15 C. Yield Changes ................ . 16 D. Growth in Production . . . . . . . . .18 IV. SOURCES OF FUTURE GROWTH . . . . . . . . . .19 A. Introduction .19 B. Lower Case Scenario . . . . . . . . . . . . . . 20 C. Sources of Future Growth . . . . . . . . . . . 21 D. Quantifying Future Growth Rates . . . . . . . . 25 E. Projections of Algeria's Food Gap . . . . . . . 27 F. Conclusions on FuturA Growth Rates . . . . . . 31 This report was written by Laurie Effron and Nicholas Krafft collecting data from a number of missions to Algeria. Susan Simonelli and Katherine Kotellos (consultants) assisted with data compilation and analysis. Philippe Ardouin- Dumazet (FAO/CP - livestock specialist) and Loretta Sonn (FAO/CP - economist) wrote selected technical background papers. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  AGRICULTURE: A NEW OPPORTUNITY FOR GROWTH Table of Contents (continuedl Page Number PART II. KEY ACTIONS: MAKING THE AGRARIAN REFORM WORK V. THE LARGER SETTING: MACROECONOMIC REFORMS . . . . . . 32 VI. TRANSFORMATION OF THE AGRICULTURAL SECTOR . . . . . . 33 A. Introduction . . . . . . . . . . . . . . . . . 33 B. The Reform.. ............. 33 C. EACs: Will They Perform? . . . . . . . . . . . 35 D. Making the Tools Available . . . . . . . . . . 39 VII. REFORMING THE INSTITUTIONS . . . . . . . . . . . . . 40 A. Introduction . . . . . . . . . . . . . . . . . 40 B. Improv:ing the Crop Extension and Research Services . . . . . . . . . . . . . . . 41 C. Improving Input Supplies and Delivery . . . . . 43 D. Liberalizing the Marketing Systems . . . . . . 46 E. A New Role for the Ministry of Agriculture . . 47 F. Livestock-Related Issues . . . . . . . . . . . 49 VIII. PRIVATE AND PUBLIC INVESTMENTS: PRIORITIES FOR THE FUTURE . . . . . . . . . . . . . . 51 A. Introduction . . . . . . . . . . . . . . . . . 51 B. Expanding Private Investment through Credit . . 51 C. The Future Role of Public Investment . . . . . 53 TX. INTRODUCING AN EFFECTIVE PRICE AND INCENTIVE FRAMEWORK . . . . . . . . . . . . . . . 64 A. Introduction . . . . . . . . . . . . . . . . . 64 B. Crop Prices . . . . . . . . . . . . . . . . . . 65 C. Input Pricing . . . . . . . . . . . . . . . . . 66 D. Livestock Pricing . . . . . . . . . . . . . . 66 E. Comparison with World Market Prices . . . . . . 67 F. The Need to Rethink Future Agricultural Pricing Policy . . . . . . . . . . 69 G. Increasing Incentives for Agricutural Exports . 71 H. Consumer and Producer Prices and Subsidies . 72 I. Impact of Macroeconomic Policy on Agriculture . 75 X. CONCLUSION . . . . . . . . . . .76 List of Working Papers and Annexes . . .78 Statistical Appendix . . . . . . . . . .79 Map No. IBRD 21310  EXECUTIVE SUMMARY PART I: ALGERIAN AGRI JLTURE: THEi PAST IN PERSPEC'CIVE AND POTENTIAL FOR GROWTH Background. Past Performance and Sources of Past Growth 1. The Government of Algeria has undertaken a series of fundamental and far-reaching reforms, aiming to increase the economy's efficiency through better use of existing capacity and placing greater reliance on the growing private sector. The new orientation focuses on: (a) developing the instruments of macroeconomic management including monetary and fiscal policy; (b) greater autonomy and flexibility of public enterprises; (c) reforming the agricultural sector; (d) restructuring the financial system; and (e) liberalizing trade, price and incentive regimes. The reforms have been most fundamental in the agricultural sector where Government dismantled the state farm sector and replaced it with about 27,000 private farms (groups and individuals). 2. The longer term objectives involve establishing a system in which enterprises are autr ,mous and competitive, the banking system acts as financial intermediary in channeling resources to the most highly profitable activities, quantitative restrictions are limited or removed, and price controls are restricted to a few "scrategic" activities or goods. The reforms underway in other sectors w'Ll thus serve agriculture by allowing a more efficient delivery system for inputs and agricultural equipment, ensuring better after-sales service and supply of spare parts, providing greater incentives for exports, and allowing a greater flow of resources to the most productive activities. 3. Agriculture is, and will remain, a key sector of the economy, accounting for 8% of GDP and 25% of employment. Mowever, food imports now account for up to 20% of merchandise imports and are growing rapidly. The growing "food gap" is a major issue for Algeria. With soil and climatic conditions similar to its neighbors, Algeria probably has a comparative advantage in a number of major cereals and some fruits and vegetables. A drive to improve productivity and increase self-sufficiency of these crops can therefore be justified on economic grounds. In addition, considerable potential exists for increasing agricultural exports to expand and diversify the export base. 4. The state farms, which accounted for about 60% of Algeria's agricultural potential, were formed immediately after Independence in 1962 on the best land in the country - - the narrow coastal strip in the north - - that had been operated by Europeans.' During the 1960s and 1970s, most of Government's investments in agriculture were directed toward this sector. The private sector comprising some Only 7.5 mn ha, or 3% of Algeria's land area is suitable for agriculture. The best land was operated by the state farm sector. The land south of the coastal strip, mainly operated by the traditional private sector, has lower and more erratic rainfall; consequently it has a higher proportion under fallow. Overall, 40- 45X of the 7.5 mn ha available for agriculture is left fallow in any year.  - ii - 700,000-1,000,000 farmers was, until the early 1980s, largely neglected. During this period, Government also carried out numerous reforms of the state farm sector: an agrarian revolution which expropriated additional land from the private sector, the formation of a new "collective-farm" sector, the creation of Government-run cooperatives to handle inputs and output marketing, and the decentralization of decision-making. Despite these and other reforms, the majority of state farms had deficits that accumulated from year to year and were kept operational only by a constant injection of Government funds to offset losses. Despite better resource endowment and access to inputs, the state farm sector performed only marginally better than the private sector in terms of yields of the major crops. Yields of private farmers were considerably higher than those of the state farms in pulses and most vegetables despite lower input use. 5. Since 1964, agriculture has grown at about 3% p.a. in constant 1986 prices, about equal to the growth rate of the population, and since 1979, at 5.5% p.a. which, by most indicators, is a good performance. While the important cereals sector has remained stagnant, the main soLrce of this growth over the past twenty years, and more especially in the last ten, has been the result of a shift into higher value, higher yielding vegetables and selected fruits and an increase in livestock production, particularly red meat and poultry. The increased supply of these commodities has been in response to increased producer prices resulting from the liberalization of marketing and prices in the early 1980s, and has been largely demand driven. In spite of this high growth rate, with a population growth rate of 3.1% p.a. and a per capita income growth rate of 1.7% p.a. in real terms, the increase in agricultural production has been insufficient to prevent Algeria from becoming increasingly dependent on imports of basic foodstuffs: the self-sufficiency ratio for cereals has fallen from about 90% in 1962 to 25% in recent years and the value of food imports as a percentage of total imports has increased from 11% in the 1970-72 period to over 20% in 19S8. The value of agricultural exports has decreased continuously since 1970-72, falling steeply from abcut 12% of total exports to less than 1% (compared to 16% in Morocco and 7% in Tunisia) in 1987, although this has been partly due to the sharp rise in the value of hydrocarbon exports. 6. Yields are low in comparison with other countries and have been generally stagnant in the twenty-five years since Independence, a major cause for concern given that input and agricultural equipment use are more intensive than in neighboring countries. The low level of yields has been primarily attributable to the dominance of socialist farms, to the weakness of supporting institutions such as research and extension, and to problems in input and equipment distribution and repair facilities. 7. Because of the relatively weak performance in productivity terms of the agricultural sector and more particularly of the state farm sector, Government instituted a series of reforms starting in the early 1980s. The principal reforms were: (a) the liberalization of the marketing arrangements and pricing of most fruits and vegetables; (b) the decentralization and restructuring of the cooperative agencies responsible for the provision of inputs and for marketing; and (c) the creation of an agricultural bank, BADR, to channel resources to agriculture. In 1987, Government undertook the most profound reform to date,  - iii - the breakup and dissolution of the 3,400 state farms, averaging about 700 ha each, into groups of about 6-7 private farmers, each group with some 80 ha. This reform fundamentally changed the structure of the agricultural sector from one with centralized control of investment allocarion decisions, productive plans and prices, to one that relies on market signals and incentives. Although the two cropping seasons following the reform, 1988 and 1989, have been difficult ones because of the profound change in ownership patterns compounded w4:h two successive years of drought, the reform has laid the foundation for a more responsive and dynamic agricultural sector which could, if the appropriate strategy is followed, lead to substantial growth in the future. Future Growth 8. Government's agricultural strategy can significantly influence future growth by taking full advantage of the reform already in place. This report has identified four specific sources of growth for the future: an increase in productivity and yields in existing agricultural areas; a decrease in land left to fallow so that the area under cultivation can be increased; increases in the area under irrigation; and development of the South. The increase in yields could comprise two parts. First, a jump to bring Algerian yields in line with those of Morocco and Tunisia, requiring increases of 20-70%. And thereafter, further increases of up to 3% p.a. which would equal the impressive growth rates achieved in Turkey. A decrease in fallow involves developing suitable technical packages and appropriate incentives. Increasing the area under irrigation and developing the South both involve substantial public investments. These sources of growth can only bc devclcped if a number of actions and policy measures is taken by Government, including improving the input and equipment supply and servicing networks, strengthening the planning capacity for public investments, and developing research and extension services. 9. Unlike the past, where growth was essentially demand driven, the projections on both population growth and per capita income increases suggest a slowdown in the increase in the demand for food. Future growth will therefore be more geared to import substitution or increased agricultural exports. 10. Three growth scenarios have been identified, using different assumptions on investment rates, policy changes and rate of institutional development. These scenarios are: - the lower growth case in which no new major actions are taken by Government. Irrigation is projected to increase at a modest 4,000 ha/yr, and the South to be developed more gradually. Policies and services are projected to remain as they have been, and inputs would be in relatively short supply. As a result, the recently privatized farms may not be able to realize a significant increase in yields. The result of this situation would be a growth rate of less than 3% p.a. between 1990 and 1995, compared to 3% over the past 20 years, and 5.5X over the past 10 years.  - iv - a middle growth case with policy changes to ensure a better distribution of inputs, equipment and spare parts, resulting in gradual yield improvements to the year 1995; establishment of extension and research services that enables yield growth at 1Z thereafter; development of incentives as well as environmentally sound farm management practices that would encourage a gradual reduction over the next twenty years of about 25X of the area under fallow; investments in irrigation at a rate of 12,000 ha per year, and development of the irrigation in the South at an additional 4,000 ba per year. This scenario gives a growth rate of 4-5X p.a. for the 1990-95 period, and is considered to be the most realistic of the three, as it takes into account the possibilities and the constraints of Algerian agriculture a higher growth case in which Algeria: (a) moves rapidly in ensuring adequate input supply and ir reforming the input delivery system so that much of the yield gap with neighboring countries can be closed by about 1993; (b) quickly establishes an effective research and extension system so that yield increases are maintained at 31 p.a. after 1992; (c) increases the irrigated area by 20,000 ha/yr; and (d) establishes technical packages and incentives so as to reduce land under fallow by 1.3 mn ha., or about 451 of the area, over a twenty year period. The result is a growth rate of 6-7X p.a. for the 1990-95 period and should be considered the upper limit of what Algeria can expect to achieve in the near to medium term. 11. Under optimal conditions, Algeria could reduce dependence on imports for selected items and modestly diversify exports. For wheats, the self-sufficiency ratio could be as high as 331 in the best case scenario or only 171 in the lower case, compared to 26X in the 1983-87 period. The projected food import bill in the year 2000 is 141 of the total merchandise imports in the higher growth scenario and 161 in the lower growth scenario, the difference representing some US$ 600 million (in current terms)in foreign exchange.' The relative insensitivity of the results on the proportion of imports to be used for food is due to the fact that about one third of the food import bill is for items not produced in Algeria (such as sugar and coffee) and because much of the growth projected here would be from fruits and vegetables which are not now imported. This underlines the importance of tackling issues that deal with both the demand and supply side of the issue of food self-sufficiency. While agricultural growth could lead to increased exports, this is difficult to quantify, as it depends on finding and satisfying markets with suitable quantity, quality, timing and packaging of the products. '/ This estimate excludes imports of maize, barley, and other agricultural items used as inputs, primarily for livestock. It should be noted that Algeria's food imports currently represent over 201 (in 1988), and up to 301 (in 1989) of total merchandise imports, but these were exceptionally poor years for agricultural production.  II. PART II: KEY ACTIONS: MAKING THE AGRARIAN REFORM WORK 12. Part II of the report addresses the issues that Government will need to resolve to tap the four sources of growth discussed above. Economy-wide reforms that directly or indirectly affect the performance of agriculture are discussed in Chapter V and include the need to liberalize regulations governing private investment, increased autonomy and reduced protection of national enterprises, liberalization of prices, an exchange rate reflecting the opportunity cost of foreign exchange to the economy, and a ma: et-based allocation of foreign exchange; these reforms would serve to imt,ove farmers' access to better equipment and spare parts and to increase incentives to agricultural exports. In addition, a number of actions aimed specifically at agriculture can be taken. These would serve to develop those commodities for which Algeria has a comparative advantage. These actions fall into four categories: those relating directly to the breakup of the state farm sector, to institutional reform, to the role of private and public investments, and to pricing and incentives. The following suzmmarizes the recommendations contained in Chapters VI-IX of the report with references to the main text. Transformation of the Agricultural Sector (Chapter VI and Volume II, Annex 2) 13. The poterttial for agricultural growth in the near term is closely related to the impact on yields of the transformation of the socialist farm system into a nrivatized svstem; estimates of growth rates are sensitive to the assumptions on the ability of the privatized farms, Exploitations Agricoles Collectives (EACs), to increase yields in a short time. The speed and magnitude of the reform involves considerable risks for the near future. These include the managerial capacity of the EACs, where the farmers are having to adjust from being wage-earn-ers to entrepreneurs (para. 6.06). Other risks relate to the availability of inputs (para. 6.15), the level of indebtedness of farmers (para. 6.14), the ability of Government to strengthen support services (para. 6.16 and Chapter VII), and the possibility of a major shift in cropping patterns, overproduction of certain crops, and a sharp drop in prices (para. 6.17). The refora took place in the midst of the 1987-88 crop season, which was also a drought year; the following 1988-89 season was also a poor year for rainfall. So it is still difficult to assess the short term impact of the reform on production and yields. It appears, however, that EAC farmers are working harder than ever, that yields in irrigated areas are up and that the potential benefits from increased productivity far outweigh the risks. Nevertheless, the reform needs to be monitored closely (para. 7.23). 14. The law setting up the EACs requires that they be farmed collectively and not divided into smaller units. Experience in other countries suggests that flexibility in the application of this law would be advantageous, as collective decision-making frequently results in sub-optimal investment decisions. The legislation should be relaxed to allow subdivision of EACs on a case-by-case basis (para. 6.09). Government should also consider the creation of a market in land-use rights which would permit farming on a larger scale in those cases where economies of scale dictate and would also allow the purchase or rental of  - vi - additional land by more successful farmers. ihis is likely to encourage farmers to develop fully their land. This should be considered in conjunction with a ceiling on the holding of land use rights (para. 6.13). The decisions on farm size and the creation of a market in land or land-use rights should be made in the context of an overall land-use policy (para 6.12). 15. EACs have been established at an average size of 80 ha, a size thought to be farmed efficiently by a tractor; hence their indivisibility. It is not evident, however, that all farms need to be this large to be farmed by suitably adapted farm equipment. Indeed, labor-intensive crops can often be farmed most productively on small farms. In addition to allowing subdivision where there are diseconomies of scale, development of private rental markets for equipment should be encouraged to permit full utilization of equipment where farms are too small to justify major investments in equipment (paras. 6.09 and 6.11). 16. Algeria's history of major agricultural reorganization is such that fac-mers may not be convinced that the current right to use the land will be any more permanent than previous reforms. If so, investment in land improvements will be sub-optimal. In the absence of title to their property, farmers need to be assured that usufruct rights are guaranteed for at least 15 years. Also, the provisions of the law should be clarified to state. except where new irrigation schemes have already been identified, that no EAC farms will be subdivided by Government fiat (para. 6.12). Reforming the Institutions (Chapter VII) 17. Changes in institutions supporting agriculture will have a rsjor role to play in enabling farmers to realize their potential and, in the lo-ger term, to expand that potential. The ability of the supporting services to deliver technical packages, of the banking system to provide credit, and of the marketing networks to deliver the inputs will determine in the longer term at what pace growth can be realized. Chapter VII focuses on the importance of the extension and research services, on the need to improve support services for input supplies and marketing, the new role for the Ministry of Agriculture, and on issues in the livestock subsector. 18. Improving the CroR Extension and Research Services. Government should simplify and clarify the extension organization, divest staff of non-extension tasks, improve the command structure and provide strong technical support for the extension workers (para. 7.06). Government similarly needs to streamline the research effort and rationalize the work programs of the research institutes. Along these lines, Government closed more than 20 research stations in 1987 and 1988 and reallocated resources to the key effort underway. To strengthen the organizational framework, INRAA's role as the central coordinating agency should be strengthened by putting into effect a new statute already drawn up; extension-research linkages should also be strengthened. Finally, terms and conditions of research service should be revised to provide an adequate career path for technical staff (paras. 7.07 and 7.08).  - vii - 19. Imnroving Input Supplies and DeliveU . Since Algeria's state fatms hed relatively good access to inputs, shortage3 of inputs were not, until recently, a major constraint. With the break-up of state farms and the creation of many, smaller units and with increased incentives to use inputs, the demand has likely increased. For the success of the agricultural reform, Government should ensure that sufficient foreign exchange is available to import agriculture inputs. The means to improve the quality and delivery of seeds also needs to be examined further. A fertilizer study is being carried out in 1990, financed by UNDP, with the World Bank as Executing Agency, to examine ways of improving distribution and to make recommendations on pricing policy. It would improve efficiency if EACs could purchase inputs from the least costly source, and private traders could stock and sell inputs, including inputs for livestock, at margins thLat make them attractive. In this respect, formerly state-run "cooperatives" are being converted to farmer-owned cooperatives. It is recommended that Government allow the full privatization of the cooperatives (para 7.11). It might also increase efficiency if enterprises other than the state monopolies were allowed to import inputs (para. 7.12). 20. Mechanizatiot1. As a result of a deliberate policy to distribute equipment to the state rarms, Algeria's official statistics indicate that it is relatively well-endowed with tractors and combine harvesters compared to its neighbors. Because of problems of spare parts and repair facilities, however, the data may not represent the real availability of functioning equipment. The future ability of Algeria to continue to mechanize depends on Algeria's industrialization policy, which currently provides absolute protection to Algeria's state industries. To allow farmers to purchase appropriate equipment at or near world prices, it should be manufactured or imported by efficient industries. It is therefore desirable to: (a) cnsure sufficient supplies of spare parts by ensuring availability of foreign exchange to importers; (b) liberalize the purchase, sale, and distribution of new equipment and spare parts at margins that encourage the private sector; (c) set prices of spare parts at a level that encourages repair of old equipment rather than its replacement by now equipment; and (d) encourage farmers and small scale entrepreneurs to set up equipment hire services (para. 7.15). 21. Liberali.i pg the Marketing Systems. Marketing is the responsibility of a plethora of bureaus and cooperatives, which should be rationalized. More recently, with the liberalization of marketing arrangements of fruits and vegetables, the private sector has begun to play a significant role in marketirg these crops (in addition to their role in ailk and meat marketing). It is recommended that Government consider further liberalization of marketing arrangements, especially for barley and other cereals produced for the livestock sector, and for export crops. The decrees on private investments should also be reviewed as should the complex system of licensing private traders and investors (para. 7.18). 22. A New Role for the Mlnistry of Agriculture. In the past the role of the Ministry was oriented toward planning, supporting and monitoring the state farms. The near term work program of the Ministry might now include: (a) building up and maintaining a viable research and extension service;  - viii - (b) strengthening its planning capacity to establish sectoral objectives and priorities; (c, creating a system of pxices and incentives that sends signals to farmers and maximizes economic returns to the country, and, if required, maintaining control of a limited number of what Government considers to be strategic crops (hard wheat, in particular); (d) building up its review capacity to screen all public investments of a certain minimum size for technical, financial, economic, social and environmental acceptability; (e) undertaking a public sector investment/expenditures review to drop all low priority projects; (f) reviewing existing input-surply and marketing arrangements to make recommendations to improve their efficiency and reduce the burden on Government; (g) strengthening its capacity to monitor the performance of the agricultural sector, including data collection and analysis; and (h) strengthening its capacity to monitor, analyze and recommend prices for those crops that are still controlled to provide adequate incentives for farmers (paras. 7.19-7.23). 23. Livestock. Responding to high milk and meat prices, livestock performance, especially poultry, has generally been good. This has encouraged the import of exotic animals for milk production which raises a number of technical issues that need further study (para. 7.25). As with crop services, livestock extension services need to be strengthened and private and cooperrstive veterinarians should be promoted to reduce the burden on Government. Finally, both the private sector and cooperatives should be encouraged to set up milk collection centers (para. 7.26). Private and Public Investments: Priorities for the Future (Chapter VIII) 24. Exoanding PXivate Investment through Credit. In addition to gzowth rates engendered by a jump in farmers' productivity from more efficient use of resources and inputs, a second major source of growth is investment, either through private means or through public investments. Government has moved away from centrally planned public investments towards relying on private investments. As a result, availability and efficiency of agricultural credit are keys to future investments in agriculture. Aside from the dissolution of the DAS, the most important change introduced by Government is the passage of the Banking Law (1986 auid amendments in 1988) making banks, including the agricultural ban'k BA')R, autonomous. Their autonomy went into effect in early 1989 and BADR has had to deal with a large demand for agricultural credit occurring in two successive drought years; BADR has twice allowed a moratorium on loan repayments. The impact of this on BADR's financial position will have to be closely monitored. This transition to autonomy will not be easy and, to prevent BADR's immobilization, Government should be prepared to provide support to BADR, on a declining basis, for a period of up to five years (para. 8.05). Govrernment has indicated its intention to establish positive real interest rates within three years (para. 8.07). In addition, although access to foreign exchange is still rationed by bureaucratic procedures, Government has simplified licensing requirements and eased access to imports for private investors, and has removed the ceilings on the value of the investments allowed for private investors (para. 8.06).  - ix - 25. The Future Role of Public Investment. Most public investments have been decentralized to the wilaya, except for large-scale, mostly infrastructure- related projects. These centrally-planned investments do, however, have an important role to play as a source of future growth. The most important of these centrally-planned investments is irrigation. 26. Investments in Irrigation. Until recently, the area under irrigation was more or less static (at about 3-400,000 ha out of a potential of 8-900,000 ha), with new irrigation about offsetting schemes that had fallen into disrepair. The major constraints to further expansion are: (a) lack of water and competition with urban and industrial users -- an issue that will become more acute over time; (b) the high cost of irrigation development and the consequent financial burden on Government; (c) the implication of low economic profitability in those schemes with low yields and/or cropping intensities; (d) slow implementation capacity, and the absence of a clear strategy and sound investment program; and (e) the risk that water development may not be sustainable because major water sources may be lost to pollution or to reservoir sedimentation. Measures should be introduced to conserve water, including inter alia, setting realistic water charges, taking steps to recharge the aquifers by using unregulated water resources and to recycle waste water for irrigation. In addition, a long-term plan for integrated use of water sources by the end-users should be developed. For the issues relating to budgetary and economic aspects, and to address the lack of a clear strategy and long-term plan, it would be useful if the Ministry of Equipment could: (a) seek lower cost solutions than have been found in the past; (b) strengthen its overall planning capacity, including its capacity to assess the economic viability of investments in the irrigation sector; (c) make a careful assessment of the relative priority of projects; (d) draw up the ten year irrigation development plan, already agreed to under a Bank-financed project (paras. 8.21-8.26). Finally, a major issue is the growing gap between completed dams or dams under construction and the construction of associated irrigation and water supply infrastructure. Priority should be given to the completion of existing projects by speeding up planning and construction of irrigation and water supply infrastructure associated with dams. If necessary, the construction of new dams should be delayed. Several of these recommendations are already being implemented. 27. In 1985, Government established independent agencies (OPI) to be responsible for irrigation maintenance. It is Government's intention that these agencies and therefore all operation and maintenance costs be funded solely through the recovery of water charges. Charges have been increased significantly since 1985 and Government is studying the issue in detail (para 8.21). Execution of irrigation projects is constrained by lack of coordination (particularly between dam construction and construction of associated irrigation and water supply infrastructure), shortage of trained staff, and shortage of essential  materials required in the construction of large scale irrigation works. In addition to measures to strengthen planning, steps should be taken to upgrade staff skills and adopt a more feasible approach to the import of irrigation pipes (para 8.24). 28. To deal with the issue of the potential loss of water supplies due to pollution or reservoir sedimentation, greater priority needs to be given to environmental issues associated with water development, including appropriate use of fertilizers and pesticides, over-exploitation, overgrazing of watersheds, and deforestation. Monitoring and control of surface and groundwater resources should also be strengthened. All these issues have particular relevance for the South (see below and paras. 8.25-26). 29. Development of the South. For social reasons and to stem the flow to the already densely populated North, Government has recently given increasing emphasis to the development of the South, where some 7-8% of Algeria's population reside. Three approaches are being used: land distribution to private farmers, rehabilitating existing plantations in oases, and drilling deep wells for large-scale cereals production. Water resources in the South are, for the most part, non-renewable. It appears that full development of the areas already distributed to farmers (200,000 ha as of end-1989) would exceed the estimated potential water resource which could lead to serious social and conservation problems in the future. In addition, given the lack of infrastructure in the South, the costs of development are high so that economic viability is an issue. Third, the region is ecologically fragile, the soil is short in organic matter, and hot summer winds are a threat to cultivated areas. Given these environmental, cost, and other considerations, a comprehensive development plan for the South should be drawn up which takes a realistic account of the natural resources of the area and the feasibility of development; in addition, monitoring and control of the water resources should be strengthened (paras. 8.27-8.30). 30. Forestry. Forestry plays an important role for agriculture through protecting watersheds and safeguarding irrigation systems, water supplies, agricultural production, and wildlife. In addition, since Algeria currently imports some $2-300 M of wood products per annum, increased wood production can be an important source of import substitution. Forest range lands are also a major factor in Algeria's fodder production. Algeria has a considerably underdeveloped potential. To protect Algeria's soils and water catchments and to increase wood and fodder production, there is a need to strengthen the planning framework, to define priority areas for erosion control works in the watershed areas, to prepare forest management plans, and to develop management and exploitation techniques to increase productivity at lower cost. The Bank has been requested to assist in this subsector and is preparing a forestry project as a first phase in a long-term integrated development effort of forestry resources in Algeria (paras 8.31-8.32). 31. Environmental Concerns and Sustainable Agricultural Growth. Relying on the various sources of agricaltural growth implies greater reliance on more intensive use of fertilizers and pesticides, which in turn will have an impact on the environment, on the quality of the water in particular, and on the long-term sustainability of the sectorIs growth. In this vein, UNDP has financed  - xi - a first phase of technical assistance to develop and help with the adoption of technical packages for integrated pest management sc as to minimize long-term hazards. Monitoring the impact of input use, and particularly pesticide residue on agricultural exports to the EEC, will be receiving greater attention than in the past (para. 8.33). To improve the monitoring of public investments, particularly irrigation development and investments in the South, it is important that Government ensure adequate staffing, training and financing of the existing environmental unit currently located in the Ministry of Equipment; it is also important that staff in the Ministry of Agriculture be trained in environmental aspects of agricultural activities (para 8.34). To ensure that development of Algerian agriculture is sustainable in the lon.g-term, ongoing operations and new investments should be examined in the context of integrated land and water resource management plans that examine the impact oni groundwater recharge, pollution, erosion, and long-term sustainability (para. 8.35). Introducing an Effective Pricing and Incentive Framework (Chapter IX) 32. If the agricultural sector is to achieve the higher growth rates indicated above, it is essential that Government has in place a system of prices and incentives that encourages farmers to increase productivity. Agriculture has generally received favorable price treatment over the past 15 years. Cereals, pulses, wine grapes, olives and industrial tomatoes are sold at prices fixed by Government, while all others are essentially determined on the free market. For several of the "controlled" crops, the wheats and barley, a parallel market exists which handles up to 50-60% of the production volume. Crop prices were more or less static until 1974, but prices of both controlled and free-market commodities have recently risen sharply. Producer prices for most crop and livestock products are considerably above import parity prices even using a shadow exchange rate (paras. 9.02-9.03 and 9.09-9.10). 33. Following many years during which input prices were subsidized, Government has adopted a policy of bringing input prices closer to "production and marketing costs", a policy of verite des nrix. As a result, input prices have more than doubled in the past five years (while the GDP deflator increased by about 15% over the same period). Government intended to remove all financial subsidies by 1989 and it is possible to state (at the end of 1989) that, starting from the ex-factory price (which does not cover costs of production), most marketing subsidies have been eliminated. These price increases have moved input prices towards their economic prices. For inputs manufactured in Algeria under inefficient conditions, the policy of removing aIl financial subsidies, including those on production, means that farmers would be paying the costs of these inefficiencies. It would therefore be more consistent to move towards a system of linking input prices to world prices, therebv insuring that they reflected economic, rather than financial, value (para 9.05). 34. Milk and meat prices are between two and four times the world market price at the current exchange rate, partly because of restrictions on trade; these prices probably cannot be maintained in the long run in the absence of subsidies. In addition, an FAQ report suggests that the relatively high livestock prices have encouraged farmers to grow low yielding cereals for the  - xii - straw and stubble value for livestock. If internal prices were to adjust to world market levels, the modern livestock industry might collapse. Should further analysis confirm it, consideration should be given to increasing emphasis on meat production from sheep and goats, to liberalizing the import of beef, and to freezing the controlled producer price of milk (paras. 9.06-9.08). 35. All prices should be set at levels that reflect the economic costs to the country of producing import substitutes or export crops; that is, in relation to world market prices. A study is to be carried out of prices and incentives similar to studies undertaken in Tunisia and Morocco (paras. 9.12-9.16) 36. With favorable prices and a conducive institutional framework, agricultural exports can be increased, although the impact on total exports will initially be small. An analysis is needed, however, of Algeria's comparative advantage of various export crops. In addition, to provide better incentives to exporters of those crops that Algeria would like to encourage, the exporters could be allowed to retain a higher portion of t'ne foreign exchange surplus earned than the 10 they are allowed at present to retain (paras. 9.17-9.18). 37. .onsumer prices of most "essential" food items, such as bread and other wheat products, have been controlled and set at levels to provide food at low cost to consumers. While subsidies on imported wheat declined between 1982 and 1988, the subsidies for locally procured wheat have increased considerably because of the recent increase of producer prices. Based on the 1989 producer prices, Government can expect to pay up to DA 2 bn (US$250 mn) a year in cereals subsidies. This represents 2.5X of Government's current account expenditures. If the recent increase in producer prices (from DA 2,700/ton to DA 4,000/ton for hard wheat) induces a supply response, and/or there is further devaluation that would increase the dinar cost of imported cereals, the subsidies required would become a further burden on the budget. While it is important to provide basic foodstuffs at affordable prices, the drain on Government's budget would be reduced if subsidies were targeted. Consumer prices and subsidies also need to be examined in the context of the proposed study on prices and incentives (paras. 9.19 - 9.25). 38. The likely impact of devaluation would be to reduce profitability of the agricultural production through the impact on input prices. Most output prices, now determined by either internal market forces where the market is protected, or by Government fiat, would not be affected by devaluation. If quantitative restrictions on imported food (mainly meat, fruits and vegetables) were removed at the same time, however, this would counteract the impact of devaluation on producer prices by putting downward pressure on them. Since agriculture is already a profitable activity, and for fruits and vegetables a highly profitable one, this would be unlikely to cause a major shift away from farming. The other major impact of devaluation would be in the level of subsidies required, which would increase sub.:antially, unless consumer prices are increased commensurately with the rate of devaluation (paras. 9.26-9.32).  - xiii - Conclusion (Chapter X) 39. Provided that Government puts in place an appropriate policy environment, Algeria's agricultural sector has the potential to grow at 6-7% p.a. over the next five years. Alternatively, if few policy improvements are introduced, it can be expected to grow at less than 3.0% p.a. In the higher growth scenario, Algeria could realize a 33% rate of self-sufficiency in wheats, its staple food, versus only 17% in the lower case scenario. Agricultural exports could grow significantly over the present modest levels, or they could remain at less than 1% of total merchandise exports. Government's commitment to resolve the issues described in this report will determine the success of the sector in realizing its objectives. 40. The major recommendations of this report can be summarized as follows: (a) The key to success for the sector in the future will be making the new land reform work. To achieve this, it is desirable that the Government be pragmatic and flexible in applying the law setting up the EACs. Both subdivision of the EACs and land rental markets should be allowed. (b) The new farmers will require a greatly improved system of support services. To achieve this, the marketing systems for inputs and output should be liberalized, cooperatives should be taken over either by farmers or private entrepreneurs, and rental markets for mechanization services should be allowed to develop. The research and extension services also need to be strengthened. (c) If stable longer term growth is to be achieved without a major drain on the Government budget, an improved price and incentives structure must be put into place that reflects economic values. To help achieve this, a pricing and incentives study needs to be carried out. In addition, it would be desirable for the Ministry of Agriculture to establish sectoral priorities for public investments that take account of financial, economic, technical and environmental factors. (d) Since the credit system will handle much of the new investment in the rainfed areas, public investment will emphasize irrigation, development of the South, and forestry. It is important that the Ministry of Equipment together with the Ministry of Agriculture strengthen its planning capacity to coordinate the use of water resources to ensure timely construction of both dams and irrigation and water supply infrastructure, and to develop area master plans for regional land use and water development; such planning should be followed by monitoring, particularly for environmental aspects, to ensure that the development is sustainable.  AGRICULTURE: A NEW UPPORTUNITY FOR GROWTH PART I ALGERIAN AGRICULTURE: THE PAST IN PERSPECTIVE AND POTENTIAL FOR GROWTH I. BACKGROUND A. General Economic Framework 1.01 At lndependence in 1962, Algeria was predominantly an agrarian society with a limited industrial base. The basis of independent Algeria's economic strategy was the promotion of industrialization and reduced dependence on foreign investments and imports. The major elements of this strategy were the application of centralized planning to the economy, reliance on public enterprises in industry and services, an agrarian revolution which included creation of socialist sector farms, and a shift from the narrow export orientation of the colonial period to a strategy of domestic self-sufficiency. The strategy was fueled by the export of hydrocarbon products which generated national savings for investment and avoided excessive recourse to external borrowing. The strategy met with some success. The investment to GDP ratio was maintained at about 40X until the early eighties, one of the highest levels in the world, and the economy, driven by these high investment rates, grew at an average of 6X per year in the 1970s compared to 3X p.a. for other middle income countries as a group. Most social indicators also registered improvement. 1.02 These impressive achievements were, however, accompanied by serious disequilibria. Economic and social infrastructure in many sectors remained well below planned targets, hampering the efficiency of the productive sectors. Consumption rates were low; unemployment rates were in excess of 15X and have now risen to around 22X. Overall, the pace of development, while rapid, was not commensurate with the enormous levels of investment. 1.03 In the early 1980s, Government moved to eliminate some of the constraints developing in the economy with more emphasis being given to improved management and efficiency. This became all the more important as most economic indicators turned sharply unfavorable in response to the 1986 drop in oil prices when the average price per barrel of Algerian oil fell by half. Export revenues fell by 36X -- a problem that was aggravated by the fall of the dollar. The country's terms of trade fell by some 60X between 1985 and 1988. The debt service ratio (as a percentage of exports) rose from 33X in 1985 to over 60X in 1988; and this was accompanied by a sharp jump in medium- and long-term external debt. The current account, which showed a surplus of some US$ 190 million a year in the  -2 - period 1980-85, moved sharply into a deficit of US$ 2.2 billion in 1986. In 1986 and 1987, a stringent import cut-back program and reduced investments brought it into approximate balance. As a result of the cut-back, however, the valte of imports in dollar terms fell between 1985 and 1987 by 27%. This has been of particular concern to agriculture because of the sharp drop in availability of agricu'ltural inputs. The actions to limit imports also reduced overall growth substantially from 3.9% p.a. over the period 1980-85 to 1.0% in 1986, and to between -1 and -2% in 1987 and 1988. After social unrest towards the end of 1988, Algeria again increased its imports and its current account moved into a deficit of about US$2 billion; growth of GDP is projected at 2.8% for 1989. With population growth of some 3.2% p.a., GDP per capita has declined for four years in a row. 1.04 The shift in economic polic- referred to above has led to the introduction of a series of reforms that covers practically all areas of economic activity. These reforms aim to decentralize the economy, make it more market-oriented, increase its efficiency through better utilization of existing capacity, and place greater reliance on the small but growing private sector. The new policy orientation focuses on (i) developing the instrument of macroeconomic management including monetary and fiscal policy; (ii) greater autonomy and flexibility to public enterprises, (iii) reforms in the agricultural sector as described later in this report, (iv) restructuring the financial system and (v) liberalizing trade, price and incentive regimes. The first significant reform occurred in 1983 vith the breakup of about 60 large Public Enterprises into 300 smaller ones. Other reforms have followed: in 1986 Government established greater autonomy for the banking sector, and in 1987 public enterprises were granted more autonomy; the Ministry of Plan was abolished and, with it, the concept and function of a priori control and central investment allocation. In the agriculture sector, the most imporrant of the reforms is the recent privatization of the state farming sector which accounted for about 60% of the agricultural potential of the country. The paragraphs below describe the agricultural sector and the recent reform program in more detail. 1.05 The economic reforms in Algeria are being carried out with the long term objectives of establishing an economic system in which enterprises are autonomous and competitive, where monopolies are the exception rather than the rule, where the banking system plays its function of financial intermediary in channeling resources to the most highly profitable activities, and where quantitative restrictions and price controls are no longer so pervasive and integral to the system. These reforms will thus serve agriculture, directly and indirectly, by allowing a more efficient delivery system for inputs, making available at economic cost agricultural equipment, ensuring a better after sales service and spare parts availability, providing greater incentives for exports and allowing a greater flow of resources to the most productive activities. B. Agricu.ltural Overview 1.06 Agriculture is now, and will remain, a key sector of the economy, because it accounts for about 8% of GDP and 25% of employment (para 2.01). Increasing agricultural production will remain a central focus of reform measures because food imports now account for as much as 15 - 20% of total merchandise imports (para 2.03) and are growing rapidly, while much of Algeria's potential for  -3- agricultural production remains undeveloped. Without specific measures to promote production, Algeria will continue to use about 16% of merchandise imports to buy foodstuffs by the year 2000 (para 4.31) and Algeria's self-sufficiency levels in basic foodstuffs could fall even further (from 26% for wheats in the 1983-87 period, for example, to only about 17% by the year 2000 -- lower case scenario -- Table 12). It is likely that Algeria, with similar soil and climatic conditions to its neighboring countries, has comparative advantage in a number of its major cereals, so that the drive to improve productivity and increase food self-sufficiency, particularly for hard -'eat, can be justified on economic grounds (para 9.15). In addition, with a Aisible limit to the potential for hydrocarbon exports, the potential for increasing agricultural exports in order to diversify and expand the export base should be explored. Physical Characteristics 1.07 Algeria, after the Sudan, is the largest country in terms of area in Africa and the Arab-Moslem world. It covers some 2.4 million km2, divided essentially into three zones roughly parallel to the Mediterranean Coast, and divided by mountain ranges: (a) a northern strip, with relatively good rainfall, divided by the Tellian Atlas Mountains from: (b) a relatively mountainous, pre-desert steppe (one-third of the country has an altitude over 1,000 m; the average altitude is 900 m); these two zones cover 15X of the country; and (c) south of the Saharan Atlas Mountains, comprising slightly over 2 million km2, is the Sahara desert, with a harsh climate, which, apart from the oases, is not inhabited or cultivated. Most of the population, 91%, live along the Mediterranean coast on 12% of the land area; only 7-8% (or about 1.8 million), mainly nomads and semi-settled Bedouins, live in the Sahara. 1.08 Given the limited amount of rainfall in the country, only 7.5 million ha, or 3%, is suitable for cultivation', while some 40 m ha are considered suitable for grazing, forests, esparto scrub and shrubs. Because some 2.5-3.5 m ha of the land suitable for cultivation are usually left fallow, the area actually cultivated in any given year is, however, only 4 to 5 m ha, 1.7% of Algeria's total land area. The area left fallow represents 40-45% of total cultivable land, compared to 28% in Morocco and 10.5% in Tunisia. Detailed land use is shown in the Statistical Appendix, Table 2.1, and summarized in Table 1 below. / With a population of about 24 million (end 1988), this translates into a per capita cultivable area of 0.32 ha, compared to 0.35 ha per capita in Morocco and 0.63 ha in Tunisia.  -4 - Table 1: LAND USE IN ALGEK'A 1986 ('000 ha)(X TOTAL USE OF LAND Cultivable Land 7,528 3 Forest, Shrubs 4,252 2 Esparto Grass 4,463 2 Prairie 31,155 13 Land not Used for Agriculture 191.183 80 Total Land Area A38,174 100 USE OF CULTIVABLE LAND Cereals 2,871 38 Fallow 2,907 39 Fruit Trees 436 6 Vineyards 130 2 Forage 716 10 Vegetables 259 3 Pulses 156 2 Industrial Crops 24 - Other 29 _ Total 7.528 100 1.09 The 7.5 million ha of arable land can be divided into five zones, as shown in Table 2 below. Zone A is liot and well-watered and is intensively cultivated with citrus and early vegetables. Zones B and C cover 66X of che cultivable land and include the plateau areas where winters are cold, crops subject to late frosts, and rainfall fluctuates sharply. Cropping in Zone B is mainly a wheat/fallow rotation and in Zone C, where rainfall is less certain, a barley/fallow rotation. In Zone D, between the dry south and the desert, cereal cropping is irregular. Zone M, in the mountains, is well-watered but, with steep slopes, is subject to erosion. The area that can be considered of rood rainfed agricultural potential is in fact extremely limited, i.e., on the order of 1.1-1.4m ha, while the area that is of marginal and/or highly variable rainfall accounts for the vast majority of arable land. This is reflected in the increasing proportion of the arable land left fallow as one moves southward, from 20-25X in Zone B to at least 50-60% in Zone D. Irrigation currently accounts for roughly 3-400,000 ha', mainly in Zones A and B and in the southern oases, representing 4.21 of the arable land, compared to 10 in Morocco and 4X in Tunisia. 1.10 Any discussion of potential for growth and agricultural development in Algeria must confront a number of physical constraints: (a) the high proportion of land unusable for agriculture and of rangelands as yet difficult to improve; (b) the high proportion of land that has shallow, erodable soils and low organic content; (c) climatic factors, including erratic and low rainfall, cold winters coinciding with the rainy season which inhibit growth when water is available, 1/ The area under irrigation is estimated at 60,000 ha under large scale irrigation and some 240,000 to 340,000 ha under small scale irrigation, including flood-irrigation, which is highly variable.  -5 - and hot summers, requiring irrigation for crop growth'; and (d) limited availability of water for irrigation, competition with domestic and industrial use, and problems of salinity. Table 2: CLASSIFICATION OF ARABLE LAND IN ALGERIA Area Fallow Zone Slope Rainfall ('000 ha) (X) (X) A < 12% > 600 mm 331 4 minimum B < 12% 450-600 mm 1,461 20 20-25 C < 12% 350-450 mm 3,507 46 40-50 D < 12% 200-350 mm 1,100 14 50-60 M > 12% > 300 mm 820 11 na Irrigated j/ 220 3 Oasis 1/ 62 1 Total 7,500 100 Classification by zone is approximate (source BNEDER) . Area under irrigation is approximate and excludes area under flood (spate) irrigation. The agricultural setting: from Indenendence to the present 1.11 At Indepei'dence, Europeans operated about 20,000 properties covering 2.4 million ha. Algerians owned some 650,000 farms covering 4.4 million ha of arable land. The European farms covered most of the better land, often in valleys and in the higher rainfall areas within the narrow strip of northern Algeria. These farms used modern farming techniques, considerable machinery and inputs. Production was oriented toward the export market, specifically for wheat, wine and citrus exports. Land farmed by Algerians was generally of lower quality, in more hilly terrain, fragmented, in lower rainfall areas, and farmed using traditional methods. 1.12 Following Independence, the land farmed by Europeans was nationalized and consolidated to create about 2,000 "self-managed" farms, averaging some 1,000 ha each, almost 10 times larger than the average European ferm of 120 ha. These "self-managed" farms were in reality state farms implementing the centrally planned objectives provided by the Government. In the early 1970s, under the "Agrarian Revolution", Government created 6,000 collective farms, covering 1.2 million ha, from state and communal land, as well as land expropriated from absentee land-owners and from the largest private farmers. Government also established monopoly marketing agencies for agricultural inputs and outputs. '/ Algeria has one of the highest coefficients of variation of production in the world due mainly to the erratic rainfall in Algeria (Table A.9 in the Statistical Appendix).  -6 - 1.13 During the 1960s and 1970s, most of Government's investments in agriculture were directed at the self-managed and collective farms which made up the socialist sector. The sector failed to perform as expected and Government attempted numerous measures to improve productivity, including the creation of Government-run cooperatives to overcome input and marketing bottlenecks and the partial decentralization of the planning process to the wilaya (provincial) level. But farm plans and investments were still effectively determined by civil servants with little knowledge of actual needs. Most farms in the socialist sector had annual deficits that accumulated from year to year (para. 2.10) and were kept operational only by injections of new "credit" from a "banking system" which acted as a conduit for the channeling of Government funds, 96X of which were to the socialist sector. In fact, since the credit was implicitly guaranteed by Government and no penalty was charged for non-repayment of these "loans", there was little incentive to improve performance either at the farm level or at the level of the individual farm worker who was assured, in any case, of a guaranteed minimum wage. 1.14 As part of the country-wide reform of the early 1980s, the 1980-84 Development Plan proposed major reforms for agriculture. The Agrarian Revolution collective farms were abolished, either merged into the state farm sector (Domaines Agricoles Socialistes--DAS) or converted to private farms; some 700,000 ha were allocated to private farmers, reversing past tendencies to increase the size of the socialist sector. The state farms were increased in number to about 3,400 and reduced in average size to make them efficient units. Overall the total area within the socialist sector was reduced from 3.5 million ha to 2.4 million ha. The private sector, representing some 740,000 farms, then stood at about 5 million ha. At the same time, Government aimed to grant further autonomy to the socialist farms and to improve their technical competence by appointing a director for each farm. In practice, little changed, and production targets and farm plans remained centrally determined. 1.15 Government did, however, embark on other reforms as part of the 1980-84 Plan, to address a number of constraining aspects of Government's strategy, most notably: a continued, almost total neglect of the substantial private sector, inadequate producer prices, overinvestment in infrastructure and the need for reallocation of credit to private farmers, and an interest rate structure that discouraged the mobilization of private resources. During the 1980-84 period, the major reforms affecting the agricultural sector, described in the matrix found in Annex 1, can be summarized as follows: (a) marketing of most fruits and vegetables, which had been controlled by Government marketing bureaus and parastatal cooperatives, was liberalized and the prices of these commodities were allowed to be freely determined; while prices increased dramatically, increased production also followed;  -7 - (b) Government decentralized and restructured a number of official marketing agencies and cooperatives to improve the distribution of inputs and the marketing of those commodities that continued to be controlled/and or handled by parastatal marketing agencies (mainly cereals, pulses, wine, citrus, dates, and some milk production); a few private traders were licensed to market inputs (except fertilizers), at the official prices and margins, and to provide spare parts and repair services for agricultural equipment; and (c) an agricultural bank, BADR, was established, whose principal mandate was to channel Government funds as well as its own mobilized resources to agriculture, including to the private farmers. 1.16 After 1985, Government continued to introduce various reforms, which included: increases in interest rates for agriculture including a 50X increase in 1986 and a 30-40X increase in 1989 (para. 8.07), and a statement of intention to move towards positive real interest rates; introduction of programs to encourage private farmers, such as provision of credit for poultry and other livestock activity and for development of the South; a cereals intensification program, to provide inputs and technical advice to increase cereals production; the appointment of technical and accounting staff to assist the DAS, in an effort to increase their productivity and reduce their losses; a streamlining of the agricultural research network of stations and institutes; and development of a policy for the build-up of an effective agricultural extension system. While all the reforms introduced since 1980 had a positive impact in reducing the losses experienced in the DAS and in providing incentives to the private sector in agriculture, the socialist farms continued to be characterized by overstaffing, unpaid debts, continued operating deficits, and low productivity (see paras 2.09-2.13 for details). 1.17 Finally, in 1987, within a larger context of economic reform and movement towards state enterprise autonomy, Government undertook the most significant and far-reaching reform in agriculture ir. over twenty years: it dissolved the socialist farms and distributed the land and the associated assets to groups of private farmers. The attempts that had been made since Independence to establish viable state farms were finally abandoned, and the private sector given the chance and the responsibility to increase output and productivity. The break-up of the DAS, which was carried out in a remarkably short time between October 1977 and May 1988, involved the distribution of land and assets to voluntary associations of no less than three farmers (see paras 6.02-6.05 and Annex 2 for further details). The immediate impact of this profound change on agricultural production has been mixed, as the farmers and associated instituations adjust to their new autonomy, and as the sector has been affected by severe drought (Chapter VI). The drain on Government's budget from continued operating losses has, however, been stopped. In the medium-term and beyond, this restructuring should lead to a more responsive and dynamic agricultural sector, which should benefit the entire economy in terms of higher agricultural production and reduced imports, providing the sector with a new opportunity for growth.  -8 - II. PAST AGRICULTURAL PERFORMANCE A. Introdu_tion 2.01 Agriculture remains one of the dominant sectors in the economy. About half of Algeria's population of 24 million (end 1988) live in rural areas and about 25X of the labor force is still employed in agriculture. Because of unattractive prospects in the agricultural sector in the past, there has been considerable rural to urban migration since Independence with a result that rural population growth has been only about 1.2X p.a. While agriculture has benefited from Government'- investment program, it has done so on a declining basis, from 25X of total investments in the late 1960s to about 61 more recently. For this and other reasons, for example, the dramatic increase in importance of thle oil and gas sectors, the sh;ire of agriculture in aggregate value added declined from about 13X in the early 1960s to about 71 in the mid-1970s (in 1980 terms). Since then, it has remained relatively constant at 6-9X, showing that agriculture has grown at the same rapid rate as the rest of the economy (Table 3). In current terms, the share of agriculture in GDP has recently increased to 12-131. Over the 1964-88 period, the gross value of agriculture in constant 1986 terms has grown at about 3.11 p.a., about equal to the growth rate of the population, but since 1980, at about 5.51 p.a., which, by most indicators, is a good performance'. As discussed later in this chapter, however, production and yields in the crucial cereal sector, which covers 60X of the land under crops in any year, has been more or less stagnant for 25 years. Most of the growth in agriculture has resulted from a shift from low to high value crops and from an increase in livestock production. Table 3: AGRICULTURAL VALUE ADDED AND GDP, 1969-1988 1980 TERMS 1969 1978 1988 DA million Agriculture value added 6,156 10,436 17,392 GDP 83,516 149,961 205,540 Agriculture value added as X of GDP 7.4 7.0 8.5 These figures are a proxy for growth rates of value added. In the absence of detailed information on the evolution of value added, however, we use growth rates of gross value. There is also some difference in growth rates if a different choice is made for the constant price year.  -9 B. Level of Food Self-sufficiency 2.02 Given a population growth rate of 3.1% per year and a high per capita income growth rate (1.7% p.a. in real terms), the growth of agricultural production of some 3.1% p.a. over the last 25 years has not been sufficient to prevent Algeria from becoming increasingly dependent on imports of basic foodstuffs (Table 4). The growth of agricultural production has been mainly in the vegetable and livestock subse'tors, where the output has high income elasticities of demand, so that demand kept pace with or generally outstripped production increases. Production of cereals and pulses, on the other hand, stagnated. In addition, the year-to-year variation in cereals production has been considerable, the result of uncertain and/or marginal rainfall over most of Algeria's arable land (para 1.09); this variability has major implications for expectations regarding the potential for future development of a secure food supply. Tabke 4: SELF SUFFICIENCY RATIOS IN ALGERIA, 1966-70 TO 1983-87 Wheats Barley Red Milk gI Pulses Potatoes Meat _____________________---- '000 tons ------------------- 1966 i97 Domestic Output 1.238 409 83 485 38 226 Net Imports 558a 4 205 Domestic Supply g/ 1,796 87 290 Self-sufficiency (5) 70 na 95 70 na na 1976-1980 Domestic Output 1,226 4779V 117 695 52 506 Not Imports 2,:186k 3079/ 14 587 67 69 Domestic Supply g/ 3,412 784 131 1,282 129 575 Self-sufficiency (S) 36 61 90 54 i8 a8 1903-1987 Domestic Output 1 112 838 169 778 55 769 Net Imports 3,2311 315 37 1,2FPZ 125 187 Domestic Supply A/ 4,343 1,151 208 2,c00 180 956 Self-sufficiency (X) 28 73 82 36 31 80 p/ Abstracting from changes in stocks. / Net imports includes cereal products; milled products converted to grain equivalent assuming a millins conversion factor of 80S. S/ Imports of milk powder, condensed milk and cream, and cheese converted to whole milk equiv6lent (WM). All milk Lmports converted at an average conversion of 1:4 and cheese imports at 1:10 to WME. A/ For 1977-80 only 2.03 The rapidly growing food gap is a major issue for Algeria. Because of the stagnation in cereals production, and in spite of the unusually high production levels achieved in the 1985-87 period, cereal imports increased sharply from an average of 2.2 m tons per year in 1976-80 to 3.2 m tons per year in the 1983-87 period (Table 4). As a result of the growing per capita consumption combined with the rapid increase in the population, self-sufficiency ratios for cereals have dropped from 90% in 1962 to 70% in the 1966-70 period and to about 25% in more recent years. For red meat, because of its growing  -10 - importance in the Algerian diet, the consistent growth rates in production (at least 4.5% p.a. in volume terms -- see para 3.'3 and Table 9) have not quite managed to keep pace with growing demand and imports of red meat represent over 15% of total requirements compared to 5% in the 1966-70 period'. For milk and related dairy products, per capita demand has almost doubled since 1975, and this must be due at least in part to a pricing policy which has kept milk prices to consumers relatively low. Self sufficiency ratios for milk have decreased from 70% during 1966-1970 to under 40% in the 1983-87 per'od. Food imports as a percentage of total imports have increased in the last fifteen years, from about 11% to 17% (Table 5) and exceeded 20% in 1988; these figures exclude imports destined for use in agricultural production, such as maize, barley and soybean cake used in livestock feed. If these items are included, the proportion of total imports accounted for by agricultural inputs and feedstuffs over the 1985- 87 period goes up to around 23%. Unless production can be increased significantly, the food gap in cereals will double over the next 20 years and become a serious burden on the balance of payments (see Chapter IV, Table 12). Table 5: AGRICULTURAL IMPORTS AND EXPORTS AS % OF TOTAL, SELECTED YEARS 1970-72 1977-79 1985-87 ---- annual average - value in mn DA ---- Food Imports - Value 724 4,644 7,350 As % of Total Imports 11% 15% 17% Agricultural Exports - Value 590 483 184 As % of Total Exports 12% 1.6% 0.3% Source: Tables A.34 and A.35 in the Statistical Appendix 2.04 The value of food exports (primarily wine, fruits, and vegetables) has also fallen dramatically from over DA 590 m in 1970-72 to around DA 185 m in 1985-87. This can be seen in Table 5. Because of he dramatic growth in the value of 'iydrocarbon exports, agriculture now accounts for less than 1% of total exports (compared to 16% in Morocco and 7% in Tunisia). It is Government's objective to increase agricultural exports to help diversify exports and reduce dependence upon hydrocarbons. C. Yields 2.05 The absolute level of yields in Algeria is low in comparison with other countries, with the one exception of barley. This can be seen in some detail in Tables A.31 - A.32 in the Statistical Appendix and in summary in Table 6 below. These yields, which have increased either slightly or not at all since Independence, are cause for concern as Algeria has similar climatic conditions as its neighbors, is relatively well endowed with agricultural equipment and has a relatively intense use of inputs (see Table 13). The low average yields are most likely the result of four major factors: (a) the structure of the '/ Imports of red meat are subject to quantitative restrictions, and thus the self-sufficiency ratios for the recent past are not true indicators of the ability of local production to satisfy demand for red meat.  -11 - agricultural sector, notably the socialist farms and centralized planning which served mainly to reduce or suppress responsiveness to incentives and resulted in continuous losses; (b) quality and quantity of inputs and of input delivery, especially certified seeds and fertilizers; (c) effoctiveness of mechanization and the impact of the mechanization policies on farm efficiency; and (d) the weakness of the supporting institutions, particularly extension and research. These factors are discussed further below. Table 6: COMPARISON OF YIELDS OF MAJOR CROPS, 1984-88 AVERAGE kg/ha Algeria Tunisia Morocco a/ Durum Wheat 745 856 1,300 Bread Wheat 732 1,295 1,400 Barley 755 569 1,057 /a For 1984-86. For more detailed data see Tables A.31 and A.32, Statistical Appendix. D. The Dual Nature of the Agricultural Sector The DAS and the private sector 2.06 Until 1987 when the socialist state farms were dissolved, Algerian agriculture was characterized by two distinct sectors, representing the classic dual economy: one sector consisted of socialist or state farms, relatively large-scale and capital intensive, on the more fertile land in the higher rainfall areas, and the other sector of private, small-scale farmers on the more marginal land, with lower and more variable rainfall. While the exact number of DAS and the area they occupied changed from year to year, in 1987 the DAS numbered about 3,400, had an average size of 700 ha, and covered a total area of 2.4 m ha, or about 30X of total arable land. Because the farms were in the more fertile area, however, the proportion of their land left fallow was much smaller than in the private sector, so that land cultivated by the DAS represented 40X of total area cropped in any year. This sector contained about 60X of Algerian production potential. It employed about 150,000 people as full time DAS members and hired part-time workers totalling about 180,000 (full-time equivalent). 2.07 By contrast, the private, small-scale sector consisted, and still consists, of some 700,000 to 1,000,000 farmers, covering about 5.2 m ha, or 70X of t!". arable area, and has many small and highly fragmented farms. About 60X of the farms are less than 3 ha, although it is likely that many farmers operate  -12 - more than one farm'. Accounting for about 70% of the arable area and 80-85% of the agricultural population, this sector cultivates only about 60% of the area under crops because much of the land falls in marginal areas and is left fallow. In the livestock subsector, by contrast, which plays an important and growing role in the value of agricultural production (para. 3.12), the private sector dominates, with ownership of about 95% of the animals. A breakdown of land use and livestock ownership by farming sector is found in the Statistical Appendix, Table A.29. Comparative performance of the two sectors 2.08 Government central planning gave the DAS priority access to inputs, equipment, credit, and technical advice. During the early 1980s, the state farms received over 65% of the fertilizers and 75% of the number of tractors to cultivate about 40% of the land cropped in any given year. Around 1984, in an effort to redress the balance in the distribution of inputs and equipment, the proportions going to the private farmers began to increase, so that by 1986/87, state farms were receiving only about 55% of the fertilizers and less than 30% of the tractors. Details are in the Statistical Appendix, Tables A.37 - A.40. 2.09 In spite of their advantages, the DAS performed only marginally better than the private sector in terms of yields in the major crops, the cereals, and wine grapes and not as well as the private sector in the bulk of the other crops, i.e. pulses, fruits and most of the vegetables. Over the period 1980 to 1986, cereal yields averaged around 100-200 kg/ha more for the DAS than for the private sector, representing a yield difference of 15-25%. During the same period, yields of the private farmers were about double those of the DAS for pulses and were generally 20 to 40% higher for most vegetables (with the exception of tomatoes) (Table 7), and this was accomplished using about half the amounts of fertilizer per ha as those used by the DAS. While average yields are not impressive by comparison with those in other countries (para 2.05), it is nevertheless noteworthy that the sector that benefited less from access to the essential factors of production (including rainfall and fertile soils) was able to surpass the more favored sector in its performance in a number of crops. 2.10 The performance of the DAS can further be measured by their profitability. Aggregate income statements for the DAS show losses year after year until 1987, although in decreasing amounts; the majority of the DAS realized losses every year until 1987, the first year that the aggregate income statement showed a net profit. In this year, about 50% of the DAS made a loss. See Statistical Appendix, Table A.30 for details. The DAS were able to continue to operate during these years only because the Government channeled funds to the DAS every year, as subsidy and as credit. The credit generally was not repaid, and was allowed to roll over into an ever-increasing overdraft, which the Government eventually took over from BADR once the DAS were dissolved. In fact, the population census of 1966 showed that there were 531,000 farmers; in 1977, there were only 403,000; yet today the estimate is that there are about .'ouble that number of farmers.  -13 - Table 7: YIELDS OF MAJOR CROPS BY FARMING SUB-SECTOR, 1980-1986 1980 1983 1986 State Private State Private State Private k&/ha Cereals Hard wheat 745 656 665 478 914 748 Soft wheat 850 723 673 439 900 680 Barley 946 792 786 549 1,022 856 Pulses 307 658 211 426 291 621 tons/ha VeRetables Potatoes 7.1 8.3 5.1 7.3 7.3 9.0 Tomatoes 11.0 8.8 9.6 10.2 28.2 13.7 Melons 5.4 6.6 4.1 7.4 5.9 8.4 Onions 8.5 8.4 5.0 8.8 5.6 8.1 Citrus 10.0 8.9 6.3 6.8 n.a. n.a. Other Fruit 1.9 2.0 1.9 2.0 n.a. n.a. Table Grapes 2.1 3.8 2.4 4.0 2.6 4.4 Hl/ha Vineyards 18 7 13 11 11 9 For details, see Table A.27 in Statistical Appendix 2.11 in 1987, the improved financial performance of the DAS can be attributed to a combination of exceptionally good weather and higher than normal yields, to particular efforts made to intensify appropriate use of inputs on cereals, and to a 23% increase in the price of durum wheat, the principal crop grown on the DAS. In addition, efforts were made to have the DAS reduce their wage costs, and the impact of this effort also appears in the comparison of the income statements of 1986 and 1987. 2.12 The poor performance of the DAS can be explained by a combination of factors. DAS farm workers were guaranteed the minimum industrial wage, regardless of the profitability of the farm. They were not responsible for losses incurred, nor did their annual losses prevent the farms from receiving credit and subsidies in subsequent years. Half of any profits were in principle to be divided among the DAS members, but it is not clear whether this was ever actually practiced. And finally, the DAS were in any case subject to following centrally determired production plans and were not free to cultivate the crops of their choice, nor were they able to sell their fruits and vegetables to private traders at the considerably more attractive prices because they were often obliged to sell to the state marketing agency. Under the combined system of guaranteed wages and centrally determined directives, there was little reason for the DAS to respond to incentives. 2.13 It was clear from their poor performance on yields and the cumulative and continuous operating deficits that there were serious structural problems with the DAS. As the momentum of the economic reforms increased in Algeria, there was ample justification for dissolving the DAS in an attempt to revitalize the agricultural sector.  -14 - A word of caution 2.14 The transfer of the DAS resources to a privatized sector has not, however, created a single, undifferentiated private sector. The newly created groups of farmers have farms that are, on average, about five times larger per farming family than in the small scale sector, and have acquired more capital-intensive and modern operations. As successors on the more fertile land to farms that had preferential treatment in their access to seasonal inputs, they may continue to exert a political pressure for favored treatment. While the reform is sure to have a positive impact in the future on agricultural production and productivity, it has not removed the dual nature of agriculture in Algeria. Government will need to pay particular attention to how its policies affect the majority of small-scale farmers.who have to date provided much of the source of dynamism in the sector. III. SOURCES OF PAST GROWTH A. Overview 3.01 Growth rates of the value of production over the past twenty-five years, measured in constant 1986 prices, have not been highly impressive at 3.1X p.a., although since 1980 the growth rate has been much stronger, at 5.5% p.a. The high growth rate for the last ten years is partly the result of the unusually strong results in 1985, 1986 and 1987, partly due in turn to exceptionally good weather'. This performance compares favorably to that of neighboring Tunisia and Morocco2. 3.02 An analysis of the sources of past growth in gross value was carried out for this report by examining changes in area planted, yields, and production levels. This analysis shows that the main source of growth cver the entire period, and more especially in the last ten years, was due to a shift into higher value, higher yielding vegetables and certain fruits, by replacing lower value crops, such as wine and citrus, and an increase in livestock production, A cereals intensification program was aJ.so undertaken by the Ministry of Agriculture, and this too had a positive impact on production levels. Because the value of cereals production as a proportion of total value of production is rather small (at about 7X, compared to 25X for vegetables and 45X for livestock products for gross production, and probably a smaller proportion of value added), this success had a relatively minor impact on overall growth rates. 2/ Official Government figures for the three countries are for growth rates of value added in constant prices and can be compared as follows: Algeria Morocco Tunisia 1977-88 1975-85 1278-87 4.9X 2.6X 3.6X  -15 - particularly red meat and poultry production. The increased supply of these commodities was the result of increasing relative producer prices following liberalization of markets for fruits and vegetables in 1982, which induced farmers to switch resources into more profitable activities (see Statistical Appendix, Tables A.41 - A.42, for producer prices). According to the data discussed below (paras 3.04-3.13), the increase in value added was not due to increasing productivity, or improved yields; nor was it the result of investment programs in irrigation, for example, as the area under irrigation did not increase significantly over this period. 3.03 This is important for the implications for future growth: the degree of substitution between crops and livestock activities is obviously limited by the scarcity of good arable land, the limited availability of water and potential for irrigation, and the limit on the area and quality of pasture land. The sector cannot therefore continue to grow indefinitely at the pace of the recent past unless efforts are made to increase irrigation, reduce fallow, and most importantly, increase productivity and yields. A second important factor of the increased agricultural production is that recent growth has been highly demand driven. Fueled by a combination of a growing population and increasing incomes, particularly after the oil price hikes, per capita incomes increased and food consumption grew rapidly during the 1970s and early 1980s. In spite of the increased supplies, however, self-sufficiency rates continued to fall (see para 2.02 and Table 4). In the future, as a result of the projected slowdown in hydrocarbon earnings, growth in per capita incomes (estimated at 1.5X p.a. between 1990 and 1995) will be slower than in the past so that demand for foodstuffs is also likely to grow more slowly than in the past. Future increases in agricultural production will have to be more geared to import substitution or increased agricultural exports. B. Changes ip Area Cultivated 3.04 Table 8 below shows the evolution in the area under various crops from 1964/66 to 1985/87. Within the cereals, which are the dominant subsector in terms of area planted, accounting for about 60X of the cultivated area, barley has taken on increasing importance, mostly by replacing hard wheat, and to some extent by reduction of fallow (between 1964 and 1974, fallow land decreased by about 800,000 ha). The incentive to grow barley has been doubly strong: first, it is a better crop than the wheats for drought resistance, and with the high variability in the rainfall over much of Algeria's cultivated area, farmers can reduce their risk by switching to barley. Second, with the increasing attractiveness of livestock production because of the increased producer prices for red meat, barley has been in high demand as animal feed. 3.05 The area under fodder crops, mainly non-irrigated vetch oats and other dryland fodder crops (irrigated fodder accounts for a small proportion of the total area devoted to fodder production), has also increased dramatically, most particularly in the past ten years. In fact, in the last two to three years, between 1984 and 1987, the area under fodder is recorded as having increased by 60%, at least partly by reducing the fallow area.  -16 - 3.06 Table 8 also shows that the area under vineyards (excluding table grapes) has decreased steadily; this area has likely been replaced by vegetables and other fruits requiring irrigation. The uprooting of vineyards that occurred after Independence may have been partly politically motivated, as the vineyards were the symbol of the former colonial rule. Since the mid-1960s, however, the uprooting has continued, while the area under citrus trees remained constant as the incentives (i.e., the producer prices) to produce these crops weakened relative to the incentives for vegetable and stone fruit production. Table 8: AREAS UNDER MAJOR CROPS, 1964-1987 Growth Rate 1964 1974-76 1985-87 l/ 1965-87 X p.a. 2/ Cereals 2.525 3.100 2,930 0 (Hard Wheat) (1,503) (1,448) (1,024) -2.1 (Soft Wheat) (452) (693) ( 547) 0 (Barley) (530) (842) (1,230) 2.8 Fodder Crops 141 223 745 7.5 Vegetables 82 165 258 5.2 Industrial Crops 18 17 22 1.5 Pulses 62 95 160 4.0 Vineyards 334 218 112 -5.9 Olives 96 176 163 3.2 Citrus 46 49 43 0 Dates 42 71 72 2.7 Other fruits 20 81 78 7.7 Figs 33 38 38 0.9 Total area under crops 3,400 4,230 4,621 }/ Although data on 1988 exist (see Table A.2, Statistical Appendix), the areas planted in 1987/88 were probably affected by the dissolution of the DAS and are, therefore, not representative of an overall trend. Z/ Based on regression analysis. C. Yield Changes 3.07 As noted in para. 2.05, the absolute levels of yields in Algeria are low in comparison with other countries. In addition, for the entire period 1964-1988. available Government statistics indicate that yields of all crops except fodder remained unchanged. grew modestly (hard wheat and vegetables), or decreased steadily (see Table 9 below). This is the result of several factors, of which the variability of the weather is a major one, affecting neighboring countries as well. With the notable exception of soft wheat yields in Tunisia, yields of cereals in neighboring Morocco and Tunisia have shown no significant upward trend over the 1973-86 period. The agricultural growth in these two countries has come mainly from irrigation development (and in Tunisia from growth of the livestock subsector). The variability of the climate in the region introduces a risk factor, reducing both the willingness of farmers to invest in higher input use and the effectiveness of such investments and this is a critical element in assessing potential for future growth (discussed below - paras. 4.08-4.12). 3.08 Most of the plantation crops (wine grapes, olives, citrus fruit) actually suffered decreases in yields, as did the pulses, and this is likely the response to decreasing relative producer prices for these crops compared to those of  - 17 - vegetables and certain fruits. For those plantations that were not uprooted for replanting, it is likely that labor and input use decreased, if indeed these plantations were not abandoned completely. For the future, therefore, the extent to which yields will be increased on these crops will depend not only on the availability of technical packages, but most particularly on the incentives that can be offered to introduce better cultivation practices. 3.09 A major factor in the general stagnation of yields was the lack of interest on the part of the state farms to make efficient use of the increasing amounts of equipment and inputs at their disposal. This in turn was due to the lack of responsiveness to incentives: with guaranteed minimum wages and virtually guaranteed membership in the DAS, and no liability for losses incurred, incentives to make profits were weak. This was the root of their inefficiency. An analysis of growth rates of yields in the state farm sector compared to private farming sector shows that almost without exception, yield increases across virtually all crops were either higher or less negative in the private sector than in the state farm sector. Now that the DAS have been broken up into groups of private farmers, the potential for responding to incentives (availability of inputs, equipment, credit, technical advice, price signals) to improve yields is much greater than in the past. 3.10 Table 9 also shows some differences for the entire period 1964-87 versus the more recent period 1979-87. For the cereals subsector, in particular, yields have been essentially stagnant when measured over twenty-four years, although they show a slight upward trend in the period 1979-87. This is mainly due to the highly favorable harvests in 1985, 1986, and 1987. When these three years are removed from the trend, the yield growth falls back to zero. Similarly, for other fruits (stone fruits) yields show negative growth over the whole period and growth at 3.2X p.a. over the more recent decade. This illustrates, in a situation of high variability in yields and production, the importance of examining trends over a long period of time.  -18 - Table 9: GROWTH RATES OF AREA, YIELD, AND VOLUME OF PRODUCTION BY MAJOR CROP, 1964-1987 AND 1979-1987 1964-1987 1979-87 Area Yield 4/ Production Area Yield 4/ Production ------- X p.a. ------- ------- X p.a. ----- Cereals 1/ 0 0 0 0 2.0 0 Hard wheat -2.1 1.2 a -4.0 0 0 Soft wheat 0 0 0 -2.9 0 0 Barley 2.8 0 4.3 5.1 0 0 Fodder crops 7.5 3.2 11.1 12.3 -4.9 6.5 Vegetables 5.2 0.9 5.9 0 5.8 10.1 Pulses 4.0 -2.0 3.31 5.1 0 0 Vineyards (wine) -5.9 -5.5 -11.4 -11.6 -5.5 -16.6 Olives 3.2 -3.1 0 -1.6 0 0 Citrus Fruit 0-2/ -2. 2! -2.5 0 -5.1 -7.3 Dates 2.7 0 2.2 0 0 0 Other fruits 7.7 -3.3 4.0 5.6 3.2 8.8 Figs 0.9 0 0 0 0 0 Livestock products 4/ Red meat 4.9I2 5.8 Poultry meat 12.3!' 12.4 Eggs 13.11' 32.6 Milk 2. 8! 3.1 1/ The growth rate in production is normally the product of the growth in area and the growth in yield. When either area or yield shows large year-to-year variation, however, production will similarly show such variation. Unless there is also a consistently strong upward (or downward) trend, the variability will dominate, and the result will be a production growth rate that is not significantly different from zero. This is the situation for cereals, where overall production showed no significant trend, although yields did show a significant trend. _/ From 1967. v/ From 1968. A/ To and including 1988. For area and production, see Footnote 1/ to Table 8. D. Growth in Production 3.11 The changes in crop production are also shown in Table 9. Except for the crops used as animal feed, the only significant growth came from vegetables, stone fruit production (apples, peaches, plums, etc.), and to a modest degree, dates. And this in turn was due primarily to the shift into these crops at the expense of other crops. Because staple food crop production stagnated, Algeria became increasingly dependent on imports (Table 4). At the same time, as domestic consumption of traditionally exported crops such as citrus, dates and wine grew, the amounts of these crops available for export decreased even more dramatically than production levels. In volume terms, for example, citrus exports decreased from an average annual level of 132,000 tons in 1977 and 1978 to virtually no exports in 1983 through 1987. 3.12 The other major source of growth in agriculture has come from the livestock subsector, which has always accounted for a large share in the value of agricultural production. In 1986, the share of the gross value cf agricultural production coming from livestock activities (red meat, milk, poultry meat, and eggs) was estimated at 45X. Table 9 shows that poultry production in particular (poultry meat and eggs) experienced high rates of growth; the extraordinary growth in egg production in the more recent past is the consequence of an explicit Government policy, started under the 1980-84 Plan, to promote this  -19 - activity through the channeling of credit to private farmers for investment in imported equipment for capital-intensive egg production. This has been accompanied by large increases in the quantities of maize and soybean meal imported as animal feed (the annual rate of increase in the volume of maize imported between 1977 and 1987 was over 20X). To have an accurate idea of what the real growth of poultry production has been, it would be useful to have estimates of changes in value added, but this information is not available. 3.13 Red meat production has also increased steadily, in response to increased relative producer prices. Data are not available on milk yields and technical parameters of sheep and cows (fertility rates, mortality rates, and average liveweight at slaughter) over time, but comparisons of these parameters with those in Tunisia and Morocco suggest that Algerian livestock is slightly less productive than its neighbors. While a source of livestock growth in the future could come from imprcved husbandry practices, this might be difficult to achieve for sociological reasons (see Volume II, Annex 3 on the livestock subsector). IV. SOURCES OF FUTURE GROWTH A. Introduction 4.01 Agricultural growth in the future will not be as strongly demand-led as in the past, both because the population growth is expected to slow down somewhat, from 3.1% p.a. to 2.7X p.a. after 1990, and because per capita income growth is also likely to slow somewhat compared to the past'. Future increases in production should, therefore, be able to contribute more to reducing imports for some crops and lead to increased exports for others. 4.02 Assessment of the potential for agricultural growth should be predicated on crops for which Algeria has some comparative advantage. While it is not yet possible to identify in detail which crops in which zones can be produced relatively efficiently by world standards, regional comparison suggests that Algeria has a comparative advantage in its major cereals, wheats and barley, as well as in citrus, dates, olives, and vegetables. A study of comparative advantage would likely find that results varied by agro-climatic zone and by the farming system to be adopted. 4.03 Potential growth rates discussed in this chapter are based on the existing relative prices, rather than on undistorted relative prices based on a study of (marginal) domestic resource costs by crop, which would be the ideal. In the absence of a set of relative prices reflecting economic value, the growth rates estimated in the following sections should be taken as rough orders of magnitude of the range of possibil'ties for agricultural growth under different scenarios. '/ In the 1970s, the growth in per capita GNP was 3.2X; up to 1985, it was around 2.3X. In recent years, it has been negative, but is projected to be around 1.51 for the 1990-95 period.  -20 - B. Lower Case Scenario 4.04 Before turning to a discussion of potential sources of future growth resulting from Government investments, policies, and actions, it is necessary to consider what a "lower case scenario" for growth in the agricultural sector would look like. The lower case would result from a situation where  new actions were taken by Government and include: (a) continuation of present macroeconomic policies, which affect in particular both the demand for and supply of equipment and spare parts, delivery of seasonal inputs, and the incentives to produce crops for export; (b) continuation of present policies on incentives; (c) continued investments to increase irrigated area by about 4,000 ha/yr and a modest pace of development (at a rate of 2,000 ha per year) for the 185,000 ha distributed in the South as part of the program to develop the northern part of the Sahara; (d) no additional institutional changes; and (e) no major impact on agricultural productivity from restructuring the DAS into Exploitations Agricoles Collectives (EACs) (paras. 6.02-6.05); this is conceivable if the EACs do not have access to sufficient inputs, equipment, and credit. In addition, with the recent cutback in imports (para 1.03), Government may have difficulty ensuring an adequate supply of inputs. This lower case scenario would result in future growth rates lower than those of the recent past. There are several reasons for this. 4.05 First, the extent to which areas cultivated with the lower value crops can be substituted for cultivation of higher value crops is obviously limited by availability of irrigated land on which to grow the vegetables and other fruits. Thus a major past source of growth among crops, substitution of area and resources Into higher value crops, is likely to slow in the future. Second, in the livestock subsector, unless husbandry practices change through intervention on a wide scale by appropriate extension advice and improved veterinary practices, the growth in red meat production will eventually be limited by availability of pasture land. For poultry and egg production, the Government's program of actively encouraging investment in these activities through provision of credit and equipment was stopped entirely in 1987 because plan production targets had been reached. A cutback on imports would also dramatically affect this sector as it relies almost entirely on imports for inputs (feedstuff). Even a resumption of the program in later years would not be at the pace of the early 1980s (with relatively free entry, the profitability of these markets has decreased over the years, and BADR is increasingly reluctant to lend for these activities in certain parts of the country). Thus, growth from the livestock subsector is likely to slow. Third, since the past performance in cereals was essentially stagnant, there is no reason to assume that cereals production would provide a source of growth. Finally, in the absence of overt actions or policy measures by Government to change the incentives to wine, citrus, and olive production, the negative growth rates of the past would be likely to continue as farmers switched land and other resources, to the extent practical, to more profitable activities. 4.06 The result of these considerations is a projected growth rate for the value of production for the lower case scenario between 1990 and 1995 of 2.9Z p.a., increasing slightly between 1996 and 2010 to 3.0% p.a. This compares to  -21 - a growth rate for the 1979-1988 period of 5.5X p.a'. The implications of this growth rate on food self-sufficiency, on the burden on the balance of payments from imported foodstuffs, and on the budget from increased subsidy levels are discussed below (paras. 4.27-4.32 and 9.21). C. Sources of Future Growth 4.07 Both Government and the EACs can significantly influence future growth in a number of ways. First, institutional changes combined with adequate incentives could result in increased productivity and increased yields and, if the incentive structure is adequate, reduction of fallow area. Second, public sector investments could increase the area under irrigation and could develop the South, which also invol 'es expanding the area under irrigation, but is considered as a separate investment/development program. Each of these potential sources of increased growth is described below, and in each case an optimistic or high growth scenario is developed as well as a less optimistic or middle case scenario. Both of these scenarios lead to higher growth rates than the lower case scenario described above (para 4.04). Increases in productivity 4.08 The increases in productivity could come from several sources. With the dissolution of the DAS and the establishment of private group farming, incremental investment/output ratios, ICORs, which were high in the past (yields in the DAS did not respond to substantial investments), should decrease sharply as the EAC farmers now have incentives to make efficient use of all previous investments in equipment, irrigation, stables, poultry houses, and other infrastructure. They should also be motivated to improve crop and livestock husbandry techniques in an effort to increase yields. Thus changing the structure of the producing units should result in a noticeable increase in yields and/or more efficient consumption of inputs. 4.09 Beyond this immediate change in yield levels, increases in productivity in the sector as a whole would come from improved quality and quantity of inputs and equipment, including improved availability of spare parts, repair services, and contract services for farms that do not own their own equipment. A critical factor for the medium term will be the effective functioning of an extension service that is able to rely on useful results from the research network. In addition, for the long-term development of livestock in Algeria, organization of grazers groups for use of the steppes will probably be needed. Finally, efficient marketing networks will have to be established for the increased output. Improvements in the institutions serving agriculture will involve a 1/ Because of a lack of reliable data on average input/output relationships by crop, the analysis is done on growth rates of gross values. This is a valid proxy for growth rates of value added, unless there are: (a) changes in productivity of either labor or capital; and/or (b) shifts from crops with lower (higher) value added to crops with higher (lower) value added. To the extent that either of these elements is relevant, the estimates of growth of gross value may under- or overestimate rates of value added.  -22- concerted effort by Government both to introduce improved policies on how the services are delivered and to invest in training and the development of human resources within the institutions. This is discussed further in Chapter VII. In any event, however, changes in yields beyond those that may be expected in the immediate future from the privatization of the DAS will come only slowly with the gradual build-up of institutions capable of delivering the necessary inputs and services. 4.10 The potential for yield increases in Algeria is considerable. If Tunisia and Morocco are used as examples of the yields Algeria could achieve, its yields of wheats, vegetables and citrus could be increased by up to 50% of existing levels. Although the climatic conditions in Algeria are harsher than in the neighboring countries, Algeria already has a more intense use of inputs and more equipment available on a per hectare basis'. The optimistic, or higher growth. scenario supposes that all conditions are favorable for the EACs to make this leap into higher productivity quickly, that is by 1993, if the weather cooperates. If, at the same time, institutions are being established to provide technical advice to the EACs, yields could continue to increase gradually, as more EACs adopted better practices. Again, using an optimistic approach to this potential, it was supposed that after closing the yield gap by 1993, Algeria could achieve a steady yield increase across these crops of 37. per year, which would equal the impressive performance of Turkey, which has sustained this rate of yield increases for major crops over a recent fourteen year period. 4.11 The less optimistic approach would suggest that the yield jump would occur somewhat more gradually, as EACs gradually adopted better techniques and learned both to manage large farms and to work together as a group. Thus, the gap might be closed by 1995, over a five year period. This middle growth scenario, taking into account the variability of climate and a number of other institutional constraints that might affect growth, further suggests that the subsequent yield increases would also be more modest, as research and extension take longer to become effective, and as the EACs continue to experience some difficulties in collective management. A yield increase after 1995 of 1% per year would be a reasonable guesstimate for a more modest growth rate. 4.12 The potential for yield increases in Algeria is an important source of growoth for the future, as the estimates below will show. It will depend heavily, however, on the abilities of the EACs to manage, on their technical expertise, on the incentives to produce, and on their access to timely and sufficient inputs, credit, equipment, spare parts, repair services, and marketing activities. Actions needed to ensure that this is achieved are discussed in detail in Chapter VI. A 1988 FAO report on the cereals sector estimates that yields of major cereals could be doubled. While this may be technically feasible, it is unlikely that Algeria could surpass the performance of its neighbors on a consistent basis, given the number of constraints in the Algerian context.  - 23 - Decrease in fallow 4.13 A second potential source of growth is a decrease in the area under fallow, which now represents about 40% of what is considered to be cultivable land. The area under fallow increases as a percentage of arable land as one moves south and as the area and climate become increasingly harsh. The greatest potential for fallow reduction in terms of total area is in Zone C (para 1.09 and Table 2), where the rainfall is marginal at 350-450 mm/yr and uncertain. Under such conditions, the inclination of farmers to reduce their fallow area would depend on the availability of seeds for improved varieties of crops such as barley and medics or new crops such as triticale that are drought resistant, and on the incentives and know-how to use them. Most of the crops that could be introduced to encourage fallow reduction would serve as livestock feed, so their attractiveness would depend on the incentives to increase livestock production. Land tenure problems may play a role in the potential for fallow reduction, as livestock owners may not be owners or occupiers of the farms and may be content to graze, or to overgraze, the pastures. 4.14 Thus decreases in the area under fallow face constraints that are technical, economic, and social in nature. Government may have some control over the first aspect, the ability to introduce suitable crop varieties for drought-prone areas, but it should not be overly optimistic on the potential for reducing fallow on a large scale in a short time. Under the optimistic or higher growth case scenario, with all technical, economic and sociological constraints reduced to a minimum, it was considered that a gradual reduction of up to 1.3 mn ha of fallow, which represents about 45% of the total area currently fallow, could be achieved by the year 2010; under the middle case scenario, it was considered that only 750,000 ha of fallow reduction would be achieved. Certain environmental concerns may be relevant to reducing fallow, particularly the threat of overgrazing on the remaining fallow, and the consequent damage to watershed areas, and these would need careful evaluation before fallow reduction programs were introduced. Increases in the area under irrigation 4.15 The third potential source of growth, and one that would result from direct public investment (as opposed to farmer initiatives) involves increases in the area under irrigation. This would result from development of new irrigation schemes and rehabilitation of existing schemes, where i:rigation can no longer be effectively used because of deterioration in the infrastructure. Algerias potential irrigated area, based on water availability and estimates of area suitable for irrigation, is estimated at 8-900,000 ha, including both large-scale schemes (LSI) and small-scale (SSI) schemes, compared to an estimated actual area under irrigation of 3-400,000 ha (60,000 ha LSI and 240-340,000 SSI). 4.16 Government has plans to develop additional irrigation areas at a rate of 20,000 ha per year, starting around 1990. This would be the ontimistic pace for development, but not out of reach if a number of technical, logistical, economic, and budgetary conditions were met. A more conservative estimate, used as input into the estimates of the ml  scenario, supposes a rate of development of 10,000 ha per year. In quantifying the growth that would come  -24 - from this development (paras 4.21-4.26) it was necessary to estimate the number of hectares that would be developed (rehabilitated or expanded) each year, the changes in cropping patterns, and the increase in production. The details of these estimates are in Working Paper 1. 4.17 Expansion of the area under irrigation would have several effects. It would provide an element of drought-proofing, thereby stabilizing the large fluctuations in production that characterize the agricultural sector at present. Since some 30-40% of the gross value of agricultural production come from the 3-400,000 ha currently under irrigation, this is obviously an important factor in stabilizing production. Second, cropping intensities would increase, thereby increasing the area under cultivation. Third, a shift to higher value crops would occur (out of cereals into fodder and vegetable production). And finally, and perhaps most import,.nt, an increase in yields would result. For the purposes of quantifying these elements, the specific estimates of increased production were taken as an average of the estimates made for the Cheliff and Mitidja Irrigation projects (Working Paper 1). Irrigation expansion also involves a number of issues touching on technical, logistical, economic, budgetary and environmental considerations, that are discussed in some detail in Chapter VIII. Development of the South 4.18 Finally, the fourth source of growth could come from development of the South, a program on which Government places a high priority for a number of reasons. It is an area sparsely populated, with both land and water resources available to be developed. The area could serve both as a pole of attraction for the densely populated North and as a source of food supply for the import-dependent country. The development program involves public investment to rehabilitate and expand areas in the Sahara currently under irrigation within existing oases and allocation of subsidized credit to individual farmers to develop currently unused land, not in the oases, by investing in wells and irrigation equipment and undertaking land reclamation. Finally, it may involve the development of relatively large-scale cereal farms, which has been explored on a pilot basis, the organization and management of which are not yet planned in detail. All the development of the South depends on making use of several aquifers that are shared with Tunisia. The program involves a number of serious and complex issues similar to those raised for irrigation, and are discussed in paras 8.27-8.30 below. 4.19 Since the development would rely entirely on irrigation, areas that are now not cultivated at all could realize cropping intensities that exceed 100%, provided sufficient water of acceptable quality can be found and exploited without damage to the water supply. For the rehabilitated areas in the oases, it would involve an improvement in yields, mainly for date palm production, and an increase in cropping intensity (date palms are often intercropped with vegetables). To what extent the increased and improved areas in production would be devoted to cereals would depend mainly on the incentives and marketing arrangements available, but experience suggests that cereal cultivation in these area.s is unlikely to be economically viable on a large scale basis. It is likely that in the foreseeable future most of the incremental production from developmert of the South would be dates, vegetables and some fodder.  - 25 - 4.20 Estimating the rate of development of the South is difficult, since no overall plan of development has been drawn up, and the technical Ministry has not yet set targets commensurate with the assured availability of water supplies as it has for development of irrigation in the Northern part of the coantry. The oRtimistic scenario could include expansion of new areas in Government financed irrigation schemes at a rate of 4,000 ha per year starting in 1995, after adequate development planning and a monitoring system had been put in place; rehabilitation could begin earlier, in 1992, at a rate of 1,000 ha per year. Because of the determination of Government to develop these areas, the middie case scenario could still involve an expansion rate of 3,000 ha per year, and 1,000 ha per year for rehabilitation. The program of distributing land and low-cost credit has been underway for some time, and as of erid-1988, roughly 200,000 ha had been distributed. Little of the land is being developed and unless an aggressive program of credit distribution and technical assistance is provided for land preparation and for digging wells, the rate of development may be quite slow. However, major risks of overpumping scarce water supplies exist (para 8.28). Under the most oRtimistic scenario, it could be envisaged that up to 4,000 ha would be put into production each year by the private sector. With less aggressive efforts, about 2,000 ha might be brought into production directly by private farmers. D. Quantifying Future Growth Rates 4.21 The effects of realizing the potential for growth from the various sources discussed above were translated into growth rates of agricultural production for the period 1990-1995 and 1996-2010 and examined for the implications of food self-sufficiency and of the size of the import bill up to the year 2000. The results for the growth rates are presented in Table 10. The lower case is an adjusted extrapolation of the past, taking into account affects of reduced imports and restricted investments in certain subsectors. The higher case scenario results from using the most optimistic assumptions on what could happen under the best possible circumstances, and the middle case scenario takes into account the various possibilities for growth, given the recent reforms and intended Government investments, but under somewhat more constrained circumstances than in the most optimistic scenario. Details of the assumptions and calculations for each source of growth and the impact on various crops and livestock production are in Working Paper 1.  -26 - Table 10: GROWTH RATES OF AGRICULTURE UNDER VARIOUS SCENARIOS 1990-1995 1996-2010 _- -- - _ ----------- p.a. Lower case scenario 2.9 3.0 Middle case scenario: all uources 4.6 3.5 Contribution to growth of: (S) (2) (1) YLeld increases 51 17 (2) Fallow reduction 8 9 (3) Irrigation investment 41 48 (4) Development of the South - 26 Higher case scenario: all sources 6.6 4.7 Contribution to growth of: (X) (X) (1) Yield increases 56 27 (2) Fallow reduction 9 5 (3) Irrigation investment 35 55 (4) Development of the South - 13 4.22 Under the higher growth scenario, the growth rate of agricultural production would be 6.6% between 1990 and 1995, which should be considered the upper limit of growth that Algerian agriculture could expect to achieve if all conditions combined in an optimal fashion. The middle case scenario of 4.6% could be considered the more realistic possibility, as it takes into account the impact of existing constraints in the sector. Finally, the lower case scenario of 2.9% over the next five years can be considered the lower limit of growth that can be expected, given the recent reforms and Government's intention to continue to build on these reforms. 4.23 Irrigation development and yield increases are the two largest contributors to the growth rates in the 1990-1995 period, while development of the South and fallow reduction contribute less (Table 10). The results on the development of the South are partly because the speed anid scope are quite limited, and partly because Government's irrigation program was assumed to start in earnest only after 1995, to allow for sufficient planning and build up of institutions to monitor the development. 4.24 For years beyond 1995, the overall growth rate will probably slow somewhat, except in the lower case, where the limits on switching to higher value crops and on available grazing lands would be partially offset by a continued investment in irrigation at the rate of 4,000 ha per year. This would allow the sector to realize a growth rate of 3.0% per year, slightly higher than projected population growth. Under the other two scenarios, growth is particularly high up to 1995 because of the impact of "closing the gap" on  -27 - yields; thereafter, growth would most likely continue at a regular, slower pace. In the later years, irrigation development makes the largest contribution to growth rates in both the higher and medium growth scenarios; this is due in part to the fact that the benefits from investment in irrigation starting in 1990 do not reach full development until four years later. Thus the full impact of irrigation development will be realized only after around 1995 anyw .y, while the impact of any large jumps in yields up to 1995 would give way to slower, steadier increases. 4.25 The program to develop the South has some impact on national growth rates after 1995 through the impact on date production. There may also be a considerable impact on vegetable availability and prices at a local or regional level. While Government hopes that the South will become something of a granary for the country, large-scale wheat production has not been foreseen in this analysis, as the incentive to grow cereals under irrigated conditions is likely to be, in the foreseeable future, less than that to grow other crops, ard the economic viability of such schemes is, in any case, doubtful (para. 8.29). Fallow reduction after 1995 has limited impact on growth rates, since greater availability of these crops may simply replace pasture grazing rather than increase meat production. 4.26 These results are based on data available in Algeria, on guess' imates as to how the EACs might respond to incentives, on assumptions as to t..e pace of Government investments, and on a number of assumptions on the response of farmers to these investments. They are orders of magnitude that serve mainly to indicate the relative importance of the various sources of growth and the upper and lower limits to growth that can realistically be achieved by Algeria. The importance of these growth rates for food self-sufficiency depends on the impact on the supply and demand of specific crops that form the staple diet for Algerians. This is examined in the next section. E. Projections of Algeria's Food Gap 4.27 To project Algeria's food gap, the impact of growth on specific crops was compared to the estimated demand for these crops in the year 2000, for each of the three scenarios. Details of the demand estimates are in Working Paper 1, and are based on projected growth rates of the population and per capita income and on estimates of the income elasticity of various products. The demand for the various crops is projected to evolve from average 1983-87 levels as follows:  -28 - Table 11: PROJECTED DEMAND FOR SELECTED FOOD ITEMS '000 tons Year 1983-87 1995 20Q0 Wheats 4,343 5,639 6,684 Pulses 180 210 228 Milk 2,060 2,330 2,752 Red Meat 206 241 286 4.28 The estimates in Table 12 below show that under the lower case situation, given the continuing high expected growth in the demand for food and, with no dramatic increase in investments or changes in policy and no perceptible impact from the DAS restructuring, Algeria will be dependent on imports in the year 2000 for as much as 83% of its wheat, 77% of its pulses, 64% of its milk, and 8% of its red meat requirements (since imports are controlled, particularly for red meat, it is likely that the real demand for red meat is underestimated here, and the self-sufficiency ratio is therefore overestimated). Since reducing dependence on imports to satisfy food requirements as much as economically possible is a major goal of the Government, it is therefore obvious that a more pro-active set of policy and actions is necessary to achieve the objective.  -29 - Table 12: FOOD GAP PROJECTIONS, 2000 '000 tons Current Ratio Lower Case Middle Case HiRher Case 1983-87 Wheats Domestic Supply 1,115 1,648 2,221 Consumption 6,684 6,684 6,684 Self-sufficiency ratio (X) 17 25 33 26 Pulses Domestic Supply 52 88 118 Consumption 228 228 228 Srlf-sufficiency ratio (Z) 23 38 52 31 Milk Domestic Supply 990 1,033 1,330 Consumption 2,752 2,752 2,752 Self-sufficiency ratio (1) 36 38 48 38 Red Meat Domestic Supply 264 267 267 Consumption 286 286 286 Self-sufficiency ratio (1) 92 93 93 82 4.29 Under the higher growth case, with increasing yields, rapid development of irrigation, and successful reduction of fallow, Algeria can modestly increase its self-sufficiency ratios above current levels for several of its major crops. This should be considered the best case scenario, and represents the upper limit of what Algeria can probably achieve over the next twenty years. Under the middle case scenario, which still involves considerable investments in irrigation and increases in yields, and which should be considered the most realistic of the scenarios, Algeria's current level of dependence on imports would be affected very little (except for red meat, which as noted above, is probably already understated because of the quantitative restrictions on imports). 4.30 A number of other factors could have a major impact on the food gap. On the demand side for cereals, a policy of reducing subsidies to consumers and increasing the final price of bread products could reduce demand and eliminate waste of bread products, thereby decreasing the need for imports. This policy would be the least costly way (and in fact, would result in significant economies for the Government's budget) to increase food self-sufficiency of the staple crops. On the supply side, changing market conditions and relative prices could result in shifting cropping patterns other than projected here and substitution, for example, of barley for wheat, or vegetables for citrus and vineyards. This might improve the self-sufficiency ratio for barley, but it would be at the expense of increased imports of wheat; it would increase the domestic supply of vegetables, leading possibly to lower prices and even some exports, at the expense of wine exports. Second, changing relative producer prices would affect  - 30 - incentives to use high-cost factors of production on certain crops, thereby affecting their yields. Diversification into other crops not explicitly considered here is also a strong possibility: these could include industrial crops such as sunflower and colza, or other oilseeds, or industrial tomatoes (these crops are at present included in the category of vegetables). These products, if produced in substantial quantities, could reduce future import levels. 4.31 The imports as projected here would continue to be a large factor in Algeria's use of foreign exchange, whichever scenario prevails. This is because of the importance within the food import bill of food items that are not produced in Algeria, such as sugar, coffee, tea, and certain cooking oils, which together account for about one third of total food imports. In the absence of pricing policy to dampen demand, imports of these items will continue to grow with the population and with incomes. The ratio of the projected food import bill in the year 2000 to the projected total merchandise import bill varies between 16X of total projected merchandise imports in the lower case scenario and 13.5X in the higher case scenario. These estimates exclude imports used as inputs into agricultural production, such as maize, barley, and soybean cakes, which could account for another 4-6% of total imports. The difference in foreign exchange savings, however, from achieving the higher growth rates as opposed to the lower growth rates, would come to about US$600 million equivalent (in current terms) on an annual basis. For details of these estimates, see Working Paper 1 in Volume II'. 4.32 In a situation where the country will need its foreign exchange to finance imports for investment, Algeria should make a concerted effort to reduce its dependence on imported food by increasing domestic production, subject to economic and other financial considerations and by using pricing policy to dampen demand. It is important to recognize, however, that policies and investments aimed at increasing domestic production will have less effect on the overall import bill than will policies aimed at decreasing demand through increased consumer prices. Given the politically sensitive nature of increasing prices of what are considered basic food items, this could only be done gradually and during a period of relative economic stability. This is discussed further in Chapter IX. '/ Algeria's food import bill in 1988 already exceeded 20% of total imports; for 1989, it was estimated at around 30% of total imports. These years were exceptionally poor for agricultural production; however, in addition, the situation was compounded by the political sensitivity of food shortages and the need to compress overall imports (the denominator) to limit the outflow of foreign exchange. This illustrates Algeria's vulnerability with respect to imported food requirements.  - 31 - 4.33 In addition to decreasing import requirements, these various growth rates imply potential for increasing exports. Quantifying this potential is more difficult, because it depends on finding suitable markets and being able to satisfy them with the quantity, quality, timing and packaging requirements. It is likely, however, that if exports of such crops as dates, olives, citrus, and early vegetables, as well as non-traditional items such as duck, turkey, and other small livestock, are liberalized and private producers, traders, and/or exporters are able to seek out their own markets, agricultural exports could increase noticeably. 4.34 Finally, these growth rates in agriculture have implications for employment and income generation and for stemming the flow of rural to urban migration, which are difficult to quantify. More intensive cultivation and higher yields imply a greater demand for labor, and the recent flow of Government employees, from urban centers as well as from local offices, to the newly created EACs testifies to ability of profitable agricultural opportunities to attract and maintain an economically active population in rural areas. F. Conclusions on Future Growth Rates 4.35 The results of these considerations show that without a pro-active policy of support by Government to take advantage of the recent reforms, increases in the value of agricultural production can be expected to be less than 3% p.a. over the next five years, considerably lower than the 5.5% growth enjoyed over the previous ten years, or it could be as high as 6.6% p.a. if a major effort is made to reduce current constraints. The results are sensitive to assumptions on the speed and ability of the EACs to realize productivity gains and on the speed, magnitude, and success of Government's intervention. Both the higher and medium growth scenarios, involving considerable response by the EACs and investment by Government, result in growth rates that exceed the projected growth vate for the economy as a whole. GDP is projected to grow at an average of 3.9% p.a. for the entire 1990-95 period, starting at 3.5% p.a. and gradually increasing to 4.2% by 1995 if the price of oil remains strong. Thus agriculture may provide an engine of growth for the economy. 4.36 In addition, these results show that if all optimal conditions combine for Government -- investment, institutional changes, and incentive framework -- Algeria could move towards its objective of reducing dependence on imports to meet food requirements and of having a modest increase in agricultural exports. These higher growth and self-sufficiency rates are critically dependent, however, on both the EACs' ability to function as autonomous collectives and on Government's ability to handle the issues facing the agricultural sector. In order to have a significant reduction in dependence on imports, furthermore, Government will have to address pricing policy issues in an effort to dampen demand and reduce waste. These issues are discussed in the following chapters.  - 32 - PART II KEY ACTIONS: MAKING THE AGRARIAN REFORM WORK V. THE LARGER SETTING: MACROECONOMIC REFORMS 5.01 Government's objectives for the agricultural sector as described in the National Development Plan are, inter alia, to: reduce dependency on agricultural imports, especially for cereals; increase rural incomes; and stem rural to urban migration. In the previous section of the report, estimates of the future potential growth rates and projections for the scope of reducing import deperdence were presented. In addition, Chapter IV presented details on the main sources of growth for agriculture that can be expected to be tapped in the coming years, and indicated that for the immediate future, in the next five years, increasing productivity of agriculture offered the greatest scope for generating growth. This second section of the report discusses potential constraints, issues, actions, and recommendations across various aspects of agricultural reforms, investments and policies that would affect the extent to which the potential for growth could be realized. 5.02 Government actions in agriculture should be considered in the light of recent past and potential future reforms, both in the agricultural sector and in the larger context of the macroeconomy. In the early 1980s, Government started to shift the economy from central planning to one that was more market-determined, although the main impetus took place in the late 1980's. These reforms translated into the passage of a banking law increasing the autonomy of banks, the granting of greater autonomy to enterprises, the liberalization of regulations governing private investment, liberalir%ation in prices and marketing arrangements in the agricultural sector, and the dramatic land reform of the state farm sector. In the future, as Government completes the shift from the centrally planned economy that has prevailed for the past twenty years, towards a more liberalized economic system, the changes will continue to affect agriculture. 5.03 On the macroeconomic front, in particular, the increased autonomy of enterprises and the reduced role of monopolies will introduce competition and greater efficiency into the services and marketing activities surrounding agriculture. Specifically, a decrease in the protection of Algerian industry, through removing absolute quantitative restrictions on imports of substitutes, will pirovide a greater range of agricultural equipment and an improved supply of spare parts. In addition, the encouragement or reduced discouragement of private enterprises should enable agro-industrial activities to expand, thereby increasing the market for agricultural products. An important change for agriculture could be continued devaluation of the dinar. This, combined with the removal of state monopoly exporters and changes in the access of private exporters to foreign exchange, should provide the incentive needed to spur agricultural exports; it may, however, also result in sharply increased subsidies to producers unless prices are realigned. Finally, reforms of the financial  - 33 - sector and introduction of autonomy and competition among banks may permit greater resources to be channelled to agriculture than was the case under the centrally planned allocation system. 5.04 While a number of actions and policy measures aimed specifically at agriculture can greatly affect the productivity and growth of the sector, macroeconomic measures will obviously play a dominant role in determining the overall environment, and ultimately the incentives and constraints that will determine the sector's long-term ability to maintain growth. For agriculture in particular, realizing the potential for growth will involve actions and policies that fall into four major categories: (a) the transformation of the state farm sector under which state farms have been eliminated and replaced by small groups of private farmers; (b) strengthening suMM,rt institutions for credit, extension and research, input supplies and marketing including, where appropriate, liberalization and/or measures to allow private enterprises to function effectively; (c) Rrivate and public investments, especially those involving credit to farmers; and (d) introducing an efficient gricing and incentive framework to provide a favorable environment to encourage farmers to increase productivity. These four areas, their impact on growth, the issues involved, and the recommendations for each, are presented in the following four chapters. VI. TRANSFORMATION OF THE ACRICULTURAL SECTOR A. Introduction 6.01 The potential for growth of Algerian agriculture in the near term is closely related to the impact of the transformation of the socialist state farming sector into groups of private farmers. As noted in Chapter IV, projections of growth rates are highly sensitive to assumptions about the responsiveness of these groups to their new situation and to what extent the needed inputs, credit and equipment will be available. This section of the report explores aspects of the recent restructuring, in particular risks associated with the reform, and, where relevant, makes recommendations for action. B. The Reform 6.02 The most profound single reform in Algerian agriculture in decades is the breakup and dissolution of the DAS, completely restructuring the means of production. In a dramatic move, Government decided in July 1987 to abandon  - 34 - its attempts -- lasting over 20 years -- to make a go of the state-owned farm sector. The decision was in line with the trend started in the early 1980s to rely more on the private sector and to give productive enterprises full autonomy, making them responsible for production decisions, profits and losses. The Government has thus introduced new incentives for producers to increase output and productivity. In the past, DAS members received a fixed and relatively attractive wage that was not related to productivity. 6.03 In December 1987, Government passed a law' providing for the division of the DAS into autonomous groups of private farmers, Exploitation Agricole Collective (EAC) and, in certain cases, into single farmer units, Exploitation Agricole Individuelle (EAI)2. The law specified that each EAC must have at least three individuals; they could be from the same affinity or family group. Where there were small parcels of land left over, generally too small for the formation of EACs, these could be allocated to individual farmers as EAIs. EAC/EAIs would be given usufrtct rights to land. EACs would operate like private farms to the extent that they would be responsible for their own planning and management. To avoid extreme parceling of holdings, EACs have been established with sufficient size to be farmed by tractors and other equipment. To ensure the future viability of these farms, the land is to be held indivisibly, although individual farmers have the right, after a period of five years, to pass their land to their children and to transfer their shares to other members. 6.04 The process of sub-dividing the DAS, forming the EAC/EAIs, and valuing existing assets for transfer to the new EAC/EAIs started in October 1987 (before the law was ?assed, the transfer was considered to be on a pilot basis). It was expected thaLt this process would take until the end of 1988. Instead, by end-April 1988 -- well ahead of schedule -- the bulk of the work had been done. About 22,000 EACs were established, each occupying an average of 70 ha of cultivable land with 6-7 members3; 5,100 EAIs have also been established, each with an average of about 6 ha. In irrigated areas, the average EAC farm size is either smaller or the number of group members has been increased; Government reserves the right, under certain circumstances, to add additional members to EACs, for example, when dryland farms are converted into irrigated farms following Government investment in irrigation. 6.05 The sheer speed and magnitude of the change involves substantial risks, particularly in the immediate future; such a radical change in ownership and means of production is bound to involve risks. The potential benefits, however, are also great -- far greater than continuing the tinkering at the margin that '/ Loi No. 87-19 of December 8, 1987. 2/ The few remaining DAS, for example dairy farms and pilot or demonstration farms, would continue to operate as state farms. 3/ The actual area distributed to each EAC was about 90 ha. This, however, includes the area under roads, buildings, swamplands, etc. not suitable for cultivation.  - 35 - has characterized past efforts to improve the performance of the DAS. Certain issues must be resolved during the course of implementation and monitoring of the reform program will be vital (para. 7.23). These issues need to be considered in the context of a comprehensive land policy that has yet to be developed (para. 6.12). The sections below summarize the major issues that are likely to em-.rge; these are discussed in more detail in Annex 2, Volume II of this report, wihich examines experiences with land reforms in other countries to assess the extent to which future problems can be predicted and, therefore, avoided. C. EACs WillThev Perform? 6.06 Managerial Capacity of EACs. The growth rates that can be achieved in the immediate future depend more on this single issue of EAC/EAI management ability than on the other issues discussed belowt, which relate to the longer term. The extent to which the EAC/EAIs will increase productivity in the immediate future will determine whether Algeria can attain the 6-7X growth rate estimated to be the upper limit of growth for the sector. Members of the EAC/EAIs comprise former DAS members, managers and extension agents who previously worked on the DAS, and others with experience and training in agriculture. A large number of senior staff of the Ministry of Agriculture left the Ministry as did many extension staff at the province and district levels to join EAC/EAIs, which testifies to the expected profitability of farming in Algeria. However, in contrast to the 700,000-1,000,000 individual, private farmers who have been farming for generations, most of the new EAC/EAI farmers have little, or often no, experience in operating their own farms -- in farm planning, deciding appropriate cropping patterns, planting, applying appropriate levels of inputs at the right time, or in cattle or poultry operations. EAC/EAI farmers are, therefore, having to adjust from being wage-earners with no experience of managing finances, credit or repayments, no responsibility for maintaining assets, farm machinery, and livestock to being fully responsible for both management decisions and day-to-day operations. 6.07 Roughly two years after this restructuring, it is still difficult to assess the likely impact of the establishment of the EAC/EAIs: the changeover took place in the midst of the 1987/88 crop season, and official statistics show that the area planted fell dramatically (by 38X) from the 1985-87 average. It was also a year of severe drought. The following year, 1988/89, has been a drought year as well, although data are not yet available to measure its overall impact. Bank missions have noted, however, that farneers appear to be working harder than before on these lands and yields are higher on irrigated areas with sufficient water. The experience of the EAC/EAIs will undoubtedly prove to be highly variable. Some groups have been formed by well-educated, dynamic individuals who have already invested substantially in their farms and who are seeking to produce high-value crops or livestock for which they have identified niches in domestic or overseas markets (spices, turkeys, ducks, etc.). Such groups demonstrate the latent energy and entrepreneurial spirit in Algeria that could lead to the rapid development of "new" commodities. This would have a marked demonstration effect for other, less entrepreneurial farmers. At the  - 36 - other end of the scale, some groups have been formed solely from previous DAS farm workers; many such farmers are older, less dynamic, and illiterate which may hamper, for example, their ability to obtain loans. Even these groups speak positively of the change and the chance to work for themselves and their children. In aggregate, there has already been an encouraging level of on-farm investment, some of which is entirely self-financed although some farmers note a shortage of equipment and inputs. To ensure that all EAC/EAIs have the necessary support to enable them to succeed, it is important that Government: (a) ensure that there are adequate equipment repair shops and spare parts; (b) encourage the establishment of contracting services for equipment rental; and (c) establish an effective extension service, making technical and financial advice available to EAC/EAIs. These aspects are discussed in more detail in Chapter VII. 6.08 While the growth rates in the next few years depend on this single issue of EAC,'EAI ability to manage, the other issues discussed in the followinp paragraphs, concerning various aspects of the recent reform, will likely affecL the EAC/EAI investment rates and levels of efficiency and, in the medium to long term, their ability to sustain higher growth rates. 6.09 Collective Management and Sub-division of Land. The law requires that the EACs be farmed collectively. Experience from other countries suggests that farming on a fully collective basis may dilute many of the potential benefits of land reform because of the difficulties of collectively deciding on investments, indebtedness, division of labor, and allocation of profits. For example, sub-optimal investments can result if the decision to invest is made at the level of the member of a group who is willing to invest least. Experience from other countries shows that (a) production generally has increased significantly when collective farms have been converted to private plots and (b) there may be considerable pressure from farmers to sub-divide their plots. Indeed, there is, in some parts of the country, an increasing pressure to allow sub-division and the takeover of the less efficient farms by the more efficient. The issues for Government are the extent to which the"indivisibility" rule that presently prevents sub-division will be adhered to and the extent to which Government will allow concentration of ownership by the more successful farmers (para. 6.10). Experience elsewhere suggests that to maximize gains from the reform, farmers should be allowed to make their own savings, investment and work allocation decisions. Accordingly, it is recommended that consideration be given to relaxing the legislation to allow subdivision of EACs on a case-by-case basis (Anne-t 2). 6.10 Farm Size. The issue of allowing sub-division of EACs leads directly to the issue of farm size. An implicit assumption of the law is that the EACs should be of sufficient size to be farmed efficiently by a tractor; hence their indivisibility. Extensive, relatively low input cropping systems such as wheat and barley that predominate in Algeria are better suited to large, rather than small, farm situations suggesting that, in certain instances, efficiency gains could be obtained by allowing for the "purchase" or rental of additional land by the better farmers. At the other end of the scale, it is not evident that farms always need to be as large as 70-90 ha; farms of 15-25 ha could be farmed  - 37 - as efficiently as the larger farms, particularly given: (a) the availability of modern, smaller, better-adapted farm equipment, and (b) the possibility of renting farm machinery. Evidence from other countries suggests that for high value, labor-intensive crops such as vegetables, smaller farms (even as small as 2 ha) can be as efficient as larger ones. It is recommended that a flexible approach to farm size be adopted so that (a) sub-division of EACs is allowed where justified and (b) development of a land rental market ts allowed to permit farming on a larger scale in cases where economies of scale dictate (Annex 2). The policies on farm size should be developed in the context of an overall land policy (para. 6.12). 6.11 Mechanization. Allowing greater flexibility in the evolution of farm size has further implications for mechanization of these farms. To ensure tnat the newly created smaller farms have sufficient access to mechanization services, even where equipment ownership may not be efficient, it is recommended that Government encourage the development of a private rental market for equipment; and that credit be provided to collectives or to private entrepreneurs to enable them to specialize in machinery services and rentals. It is also recommended that Government strengthen its research on selection of equipment suitably adapted to Algeria's new farming conditions (paras. 7.13-7.15). 6.12 Land Policy and Permanence of Land-Use Rights. The law provides for the transfer of usufruct rights in the event of death of an EAC farmer but it is not clear whether this transfer can be to more than one farmer. Nor is it clear whether EAC members are allowed to buy out other members. It appears that, as a general rule, changes cannot be made that affect the existing farm size or boundaries. An important issue is whether the usufruct rights are permanent, and, more importantly, whether farmers are convinced that these rights are permanent. Algeria's history of major agricultural reorganizations over the past 25 years with land expropriated from private farmers, then redistributed to other private farmers, is such that some farmers may not be convinced that the inheritable right to use the land will be more permanent than previous reforms. This is a potentially important issue in Algeria. Some private farmers, whose land was taken in the 1970s during the agrarian revolution and distributed to the DAS, are asking for their land back. Why, they argue, should ex-DAS workers or ex-staff from the Ministry of Agriculture be given land that had been in their families for generations? It is recommended that this issue be addressed as a matter of urgency: experience with land reform in other countries shows that if farmers remain unconvinced by guarantees in the law that the land is theirs, investment in and maintenance of land improvements are likely to remain below optimal levela. In addition, the law leaves room for Government to add new EAC members' or subdivide EACs if, for example, Government develops a new large-scale irrigation scheme on land belonging to an EAC. Responding promptly to some of these concerns, Government has issued an interministerial circular clarifying that if EAC members develop their land by sinking a (private) '/ Article 24; "In every case, the state can exercise its right of pre- emption", and Article 25: "Except for the state..., any acquirer must be ... agreed to by other members".  - 38 - tube well, Covernment will n__ add additional EAC members. In this situation, EACs will remain the sole beneficiaries of their own investments. The situation is less clear if Government finances a large-scale irrigation scheme that is not currently envisaged in a particular region. Will Government then subdivide existing EACs thereby taking over investments already made by the EACs? Any expropriation would undermine investments in areas that farmers believe may be irrigated in the future. It is recommended that, except where new irrigation schemes have already been identified and EACs told that farms will be subdivided, Government clarify that no EAC farms will be subdivided by Government fiat'. Government also needs to assure farmers that the usufruct rights are guaranteed for at least 15 years, a period found crucial in China. Details of this issue are in Annex 2. It is recommended also that consideration be given to the introduction of a proper land market based on land titles, rather than merely a market in user-rights. If so, steps should be taken to provide farmers with titles to their properties. In this context, it is recommended that Government develop an overall land policy with clearly defined objectives so that future changes can be considered within the framework of such a policy. 6.13 Transfer of Usufruct Rights. Based on experiences in other countries, it is recommended that creation of either (a) a market in land-use riglhts or (b) a land market be allowed. The presence of such a market is likely to encourage farmers to develop the potential of their land and to assist in the transfer of poorly farmed, unproductive land to better, more productive farmers. If such a measure were considered, it could be introduced with specific ceilings on total land area over which one farmer can hold land-use rights. Other countries have also found that the development of rental markets in land-use rights can bring many of the efficiency benefits of a sales market without the same potential social costs (see Annex 2). 6.14 Level of Indebtedness. The newly established EAC/EAIs have purchased the former DAS' farm assets, machinery, equipment, livestock, and plantations from Government. These assets (called the "patrimoine") have been valued on the basis of formulae determined by Government reflecting both the cost of the asset and the fair market value. In the Mitidja area, for example, the value of the assets transferred was DA 800,000 (US$135,000) per EAC or US$17,000 per beneficiary. Credit for the patrimoine was provided by Government through BADR on a long-term basis2. EAC/EAI farmers also had to borrow for subsistence until the next harvest and for seasonal inputs. They also need to invest further to replace worn out equipment or to complete infrastructure. Many new farmers have thus started their existence with high levels of debt. Farmers may be tempted either to keep requests for loans to a minimum, thereby reducing potential '/ For equity reasons an improved cost recovery policy for nationally built irrigation schemes would, however, also need to be introduced. 2, Repayment is over 15 years, with five years of grace for interest payments and two years of grace on the principal, at an interest rate of 2 1/2X p.a. Since the funds are paid to Government (through BADR), this should not affect BADR's liquidity.  - 39 - productivity, or at the other extreme, may mishandle their financing and become over-indebted. In fact, there has been a rapid increase in the demand for credit, although the majority is for the purchase of seasonal inputs. The level of indebtedness does not, therefore, appear yet to be a constraint to future investments. Also, farmers seem to differentiate between loans from the banks, which they regard as a "legitimate" debt, and loans from Government for the patrimoine for which they take a more casual approach. These issues will need monitoring by both BADR and by the Ministry of Agriculture to ensure that a proper balance is achieved between appropriate investment levels and indebtedness and to ensure that Government debts, as well as debts from BADR, are repaid. One way to alleviate some of the indebtedness would be to extend the five year grace period on the "patrimoine". D. Making the Tools Available 6.15 Input Supply. This is probably the single most urgent issue concerning the envirornment in which EAC/EAIs must perform, and one that will profoundly affect the ability of these new units to respond to incentives in the immediate future. Many of the seeds, some finished fertilizers, inputs for the manufacture of some pesticides and fertilizers, animal feed and some of the agricultural equipment and spare parts are imported. With the cutback in imports (para 1.03), the ability of Government to supply the agricultural sector with sufficient inputs has been threatened. At present, there appears to be an adequate supply of fertilizer only in some parts of the country; in other parts, fertilizer is in short supply. Animal feed, veterinary supplies, spare parts, and to a lesser extent, agricultura'l equipment and machinery, are in critical shortage throughout the country with a particularly negative impact in the livestock sub-sector. It is vital that EAC/EAIs be adequately supplied. This is important not only for the EAC/EAIs themselves, but also for Algeria'a entire reform program, which would be set back if this, the first major reform, were to get off to a bad start. It is therefore recommended that the Government take measures to ensure that sufficient foreign exchange is allocated for the import of the key agricultural inputs and equipment required to support sector demands. 6.16 Support Services. In the past, the support services, including research and extension, input supplies, equipment sales, spare parts, and after sales service, credit and output marketing were oriented toward a small number of large producers, the DAS. In addition, delivery of most inputs, equipment and credit were determined by central planning following centrally established targets for production. The entire system has now changed and the networks of input cooperatives, equipment suppliers, and the system for credit delivery will have to adapt to a new structure and most likely an increased demand from a larger number of smaller units. In addition, EAC/EAIs are supposed to be competing on an equal basis with the small-scale private farmers for these services. This raises a number of issues of basic policy on how these services are to be strengthened or changed which are explored later in Chapter VII. 6.17 Prices. Most new EAC/EAI farmers are optimistic about profitable opportunities from farming. At present, vegetables, some fruit and flower production are highly profitable and many EAC/EAI farmers are shifting into  - 40 - these commodities. This, in turn, may lead to over-production and a subsequent decline in prices and profitability of these crops (vegetable prices have already declined somewhat), thus disappointing expectations. At the same time, and to the extent that technical considerations allow, farmers will probably shift out of cereals. Fortunately, three factors will act to minimize these risks. First, any such adjustment process would be relatively short, albeit painful, before new equilibrium levels of production and prices are established. Second, a number of prices of key staples, such as cereals, are still controlled or have floor prices established by Government, which should serve to stabilize prices and reduce the impact of major short-run fluctuations. And third, production of the vegetables and fruits is largely dependent on the availability of irrigation which is quite limited. Thus, while the extent of falling prices may be limited, the risk of some instability in profitability and farmers' response must be recognized. While it may be tempting for Government to reintroduce controlled marketing or fixed prices for those crops already liberalized, this is not recommended, as the market forces of supply and demand permit more satisfactory channeling of resources and commodities for these crops. It is recommended, however, that the role of the national fruits and vegetable marketing agency (ONAFLA) be monitored and, ir. the early years, be allowed to provide effective floor prices for selected vegetables and fruits. The larger issue of prices and incentives is discussed in Chapter IX. VII. Reforming the Institutions A. Introduction 7.01 While performance of the EAC/EAIs probably ranks as the single most important determinant of increased productivity in the agricultural sector for the near term, and therefore the most significant factor in determining growth rates, changes in institutions supporting and serving agriculture will also have a major role to play in enabling farmers both to realize their potential and to expand it. For increasing yields and reducing fallow in particular, the ability of the supporting services to deliver technical packages, of the banking system to Provide credit for the necessary investments, and of marketing networks to deliver the necessary inputs, will determine in the medium term whether and at what pace these sources of growth can be realized. Within the larger context of promoting the autonomy of enterprises and encouraging the private sector (see paras. 1.04 and 5.03), Government has made a number of institutional changes designed to improve services available to the agricultural sector. While these past reforms, presented in some detail in the matrix in Annex 1, have served to improve agricultural efficiency, now that the sector has been fundamentally restructured and centralized planning no longer serves as the motor of agricultural production, further reforms are called for. 7.02 The most important institutional reform undertaken so far is that in the banking sector. Credit has the potential to become an important engine of growth and, in the future, will be the major source of investment in agriculture. Key actions that are required to realize this potential are discussed in Chapter  - 41 - VIII which deals with private and public investments. This chapter focuses ou the institutional actions that need to be taken to provide efficient services to the farm sector. The most crucial of these is the establishment of sound extension and research services. Actions also need to be taken by Government to create an environment in which non-Government support services of input supplies and marketing (and credit) can function efficiently. B. Improving the Crop Exter,sion and Research Services 7.03 The ability of farmers to increase yields in the immediate future probably depends most on their willingness to apply techniques and adequate amounts of inputs already familiar to them. The change in their habits from the recent past would mainly be the result of the fundamental change in the incentives they have to eliminate waste and increase yields. Beyond these improvements, however, increasing yields of less well-informed farmers, and increasing yields through introducing better technical packages will depend on an effective extension service backed up by a research system responding to the needs of the farmers. 7.04 Private farmers were almost totally overlooked in terms of agricultural services until 1980. Recognizing the weakness of the agricultural extension service, the 1980-84 Plan envisaged the establishment of an extension service to assist private farmers (Delegation d'Agriculture du Daira-DAD). A beginning, albeit slow, has been made in the initiative. Unfortunately, the number of extension agents remains limited (ratios of as few as 1:4,000 farmers) and much of their time is taken up with administrative tasks. The previous shortage of extension staff has been compounded with the departure of many extension (and some research staff) to join the EAC/EAIs. Extension agents also lack transport and funds for extension. The DAS were served by higher level staff grouped into Isecteurs de developpement agricole" (SDA). 7.05 The basic tenets of a good extension system are the presence of a simple, preferably single line of command, regular farm visits by single-purpose workers, regular training of motivated staff and good links to research. The extension service also needs to be adequately funded. While much has been done to improve the extension system recently -- in 1985, Government issued a national extension policy to put in place appropriate staffing and management, improve the technical content of the extension system, improve the farmer/extension/research linkages and provide better support to the private sector -- none of these conditions is yet met adequately. Responsibility for implementation of the extension service rests at the wilaya (province) level and is based on a highly decentralized system with no line responsibility to the center. 7.06 Recommendations on Extension. The first step in establishing a sound extension system is to: (a) simplify and clarify the organizational structure that exists and (b) divest those staff who do extension from most of the administrative and non-extension tasks. Second, there is a need to develop a professional service based on a direct line of command, strict observance of working programs, close cooperation with research, strong technical support from  - 42 - teams of subject-matter specialists, and an effective training organization. To aid the planning and monitoring of this effort, Government should improve the data on the private sector which is presently weak - - it is not clear, for example, how many farms are in operation. Th'rd, the system of budgeting for extension should be changed so that agricultural extension is included as a line item in the wilaya budget with controls to ensure adequate and timely funding. Fourth, the graduate staff released from the breakup of the SDA should provide the basis for building up teams of subject-matter specialists at the wilaya level. These teams of well trained and motivated technical staff will be the key link in the transfer of existing know-how from the research stations to the field. In addition, it is imperative that technical messages be improved, including much needed recommendations on soil conservation, on-farm erosion control, watershed management, and fertilizer and pesticide use taking account of environmental considerations. Government has requested the Bank to support an extension and research project designed to establish a strong and responsive extension service. In view of the limited experience with agricultural extension in Algeria, the project would be a pilot operation to test extension methodologies which could be extended in a second phase. 7.0i On research, Government has mounted a program of reforms for making the service more effective and to redress problems of a poor and frequently changing organization, weak staffing, lack of coordination between research institutes and absence of managerial autonomy. The main features of this reform program are, first, stronger central coordination of the work of the technical institutes (para. 7.08). Second, in order to streamline the research effort and avoid duplication, research work will be organized from 7-8 Regional Agricultural Research Centers covering the main agro-ecological zones, and the number of outstations will be further reduced (some 20 stations were closed in 1987 and 1988, reducing the number to 40). Third, research priorities have been identified and a system of programming and budgeting by objectives is to be implemented. 7.08 A number of research issues remain, however. First, the Institut National de la Recherche Agronomique d'Alg6rie (INRAA) which has a coordinating role ut.,er its existing statute, has sometimes found it difficult to exercise this role against the stronger technical research institutes. Strengthening this role will be an essential first step in improving coordination. To this end, a new statute has been drawn up for INRAA. Government now needs to confirm its decision to apply the new statute and to issue the appropriate ministerial instruction. A second issue relates to the need to establish much stronger research-extension linkages with adaptive and on-farm research oriented toward farmers' constraints. This will be particularly important in servicing the EAC/EAIs. A third issue affects both the extension and research services. Each service faces difficulties due to the lack of a recognized career path for staff; remuneration levels are not competitive with other sectors. In research, a clear career path is required to attract capable scientists, while in extension, motivation of field staff is crucial. Revised terms and conditions of employment are currently under review. It is recommended that these terms and conditions be revised as a prerequisite for establishing sound extension and research services. The anticipated Bank-supported project aims to address these issues,  - 43 - strengthen the central coordination of research work, and focus on the implementation of five priority research programs (cereals, legumes, forage crops, red meat production, and cereal/livestock system). By introducing these elements of sound research and extension systems Algeria can put into place the prerequisites for longer-term growth of productivity in the agricultural sector. C. ImRroving Input Supplies and Deliver, 7.09 Ouantities. Both in the immediate future as well as in the longer term, one of the most critical elements that will permit Algeria to realize its potential from the various sources of growth will be an adequate quantity and good quality of inputs, available to all farmers. In the past, the average quantities of inRuts, certified seed and fertilizers, used per hectare in Algeria have been high compared to neighboring countries, which nevertheless have consistently achieved higher yields (Tables 6 and 13). Distribution of inputs was uneven, however, with the national marketing bureaus and local cooperatives focusing almost entirely on the socialist sector which did not use these inputs efficiently. The DAS, which cultivated about 40X of the total planted area, received 60-70X of available fertilizer. This situation improved somewhat so that by 1986 the private sector received a larger share of inputs (for details, see Statistical Appendix, Tables A.37-A.40). Tabe2L : COPBATIVE USE OF INPUTS AND EQIIPMENT, 1984/85 aMeria Tunla uQCCO Une of seasonal inputs k8/ha Use of certified cereal seed 28.3 16.5 14.8 Use of fertilisers 54.8 17.2 51.3 MebhanizgAtm ha/unit Number of bectares cultivated by: Tractors 87 166 324 Ploughs and other equipcut 40 94 n. a. Combine harvesters (cereal hectarage) 503 905 n.a. Fertilizer spreaders 1,203 1,382 n.a. Seeders 803 1,146 n.a. Pesticide sprayers 402 1,175 n.a. Source: Alerie: Benk estimates; Morocco: Agricultural Sector Adjustment Loan (World Bank 1985, Technical Annex); Tunisia: World Bank and FAO/CP: Plan directeur do mdcanisation agricole pour La Tunisie. 7.10 Oualitv and Distribution of InDuts. Cereals and pulses seeds are produced locally. There are, however, problems of quality control in seed multiplication, and in subsequent certification, storage, and distribution, which are all the responsibility of the parastatal marketing bureau and its system of Government-sponsored cooperatives, OAIC and the CCLS. One possibility for improving the supervision of seed multiplication and subsequent stages of certification and storage might be to establish a specialized agency; alternatively, OAIC and the CCLS may need additional logistical and technical  - 44 - support to carry out this function. The EEC is currently financing a pilot project to test approaches to develop the technical, organizational, and legal framework for the production of quality certified seed. It is expected that lessons learned from this operation can be replicated so that Algeria can produce more high-quality seed locally. Fertilizers and Resticides are also mostly produced locally by a parastatal fertilizer and chemical company, and distributed by a parastatal marketing agency and its network of cooperatives. The timing of the delivery and quality of these inputs need to be further explored, as does the use of certain pesticides and fertilizers, including the impact on the environment; in this regard, stricter guidelines, the ban of certain pesticides and the training of extension staff in the use of pesticides and fertilizers and their proper disposal should be a priority concern (paras. 8.25 and 8.33-8.35). 7.11 Government has carried out a review of some of the many institutions providing inputs to farmers. For the time being, no major change is expected in the role of input delivery to Government-sponsored national-level "offices," which involves mainly importing supplies. Over time, however, the "offices" would presumably become more autonomous, in line with the recent law on autonomy of enterprises. Government did pass a decree regarding agricultural cooperatives: "Decret no. 88-170", dated September 1988. The intention is to transform the existing cooperatives from "top-down" institutions run by Government into "true" cooperat'ves established by, and run for the benefit of, farmers. These new cooperatives are free to do most service-related activities in the agricultural sector, for example, to purchase inputs from the cheapest source, by importing directly if they have access to foreign exchange, purchasing directly from the local manufacturer, or purchasing directly from private suippliers if such are allowed to exist. Where no farmers express a desire to take-over or "buy" the cooperatives, a service enterprise can be set up with the right.3 of a private company after purchase of the assets. The intention is that the vreal" cooperatives be run like private companies, and, accordingly, the assets of the previous cooperatives are being sold to farmers on credit. It is not clear that 'real", bottom-up cooperatives can, in fact, be established by Government fiat and, in the absence of the significant Government support that was forthcoming to the previous cooperatives, some of the new cooperatives may well collapse financially. The less successful cooperatives can presumably, according to the law, be taken over by ex-employees, individual farmers, or private entrepreneurs and run as private enterprises. It is recommended that Government continue to encourage the formation of these "real" cooperatives but that Government's financial support be limited and should not go beyond, say, the end of 1990. It is further recommended that the Government not support unviable cooperatives and that, in such instances, it not hinder their takeover by the private sector. 7.12 Regarding input pricing, Government currently has a policy of uniform input prices throughout the country. CASSAP and its parent, ONAPSA, the agencies responsible for distribution of fertilizer, thus make larger profits on fertilizer distributed and purchased near the factories and smaller or negative margins at points far from the factory. If EACs or private traders were allowed to purchase directly from the factories, they could realize profits by virtue of their proximity to the factories. This is unlikely to be condoned. At the  - 45 - same time, it is clearly not in the interests of CASSAP/ONAPSA, which have a monopoly on the distribution of fertilizer, to encourage competition and/or the direct purchase from factories. Nevertheless, to provide inputs to farmers in a reliable manner at the cheapest possible price, competition among input suppliers is desirable. Accordingly, it is recommended that Government allow, in addition to cooperatives, private traders and EAC/EAIs to: purchase inputs from the cheapest possible source, and stock and sell inputs, including inputs for livestock, at margins that make them attractive. To this end, it is also recommended that Government liberalize the import of fertilizer and other inputs, permitting state enterprises other than ASMIDAL (which has a monopoly on importation of fertilizer) and private entrepreneurs to import fertilizer and other inputs. Finally, it is recommended that prices of fertilizers and other inputs be liberalized, at least at the retail level, to encourage enterprises and individuals to market these inputs. Government is about to embark on a major study of the fertilizer subsector financed by UNDP with the Bank as Executing Agency which will, inter alia, evaluate different options for fertilizers including price policy, distribution margins, the role of the marketing cooperatives, the role of the private sector and environmental aspects (para. 8.33). 7.13 Mechanization. As with inputs, the ability to improve the scope and quantity of equipment available to farmers, and especially to improve the spare parts and repair services available, will play a critical role in the ability of farmers to respond to incentives and to participate fully in large-scale investments such as irrigation or development of the South. As with inputs, available data on mechanization levels for Algeria show that by comparison with neighboring countries, Algeria is relatively well-endowed with tractors and combine harvesters (Table 13). This is the result of (a) the importance accorded by Government to the distribution of equipment, particularly to the DAS; (b) an over-valued exchange rate that distorts the price of capital relative to labor and encourages over-capitalization; and (c) a pricing policy that kept new equipment prices low for many years; prices have been increased recently (spare parts have always been sold at relatively higher prices). The averages in Table 13 above mask the even more intensive level of mechanization that existed in the socialist farms before their break-up and which should now exist in the EAC/EAIs that have acquired these assets. This is because private farmers were able to obtain only a small portion of the equipment distributed. Because of the problems of spare parts and repair facilities, however, these figures may not represent the real availability of functioning equipment. In addition, because much of the manufacture of equipment is based on a turnkey arrangement from the early 1980s, the equipment currently manufactured and distributed in Algeria is not always well adapted to Algerian conditions. 7.14 The effectiveness of mechanization in the future in Algerian agriculture is direcLly related to the larger issue of Algeria's industrialization policy which provides absolute protection to the nascent parastatal industry manufacturing the equipment; this issue also affects the import and manufacture of fertilizer and other inputs by existing state monopolies. To allow farmers to purchase inputs at or near world market prices, agricultural inputs should be imported or manufactured by efficient local enterprises. While Government's  - 46 - industrialization policy may achieve its objective of replacing imports of finished manufactured goods by domestically manufactured goods, it increases costs through lack of competition and/or inefficiencies in manufacturing. It also limits farmers' choice of equipment, limits the availability of equipment reflecting the latest technology, and equally as important, sharply reduces the availability of spare parts and after sales service. The monopoly state enterprise responsible for importing and maintaining a stock of spare parts has been unable to cope with the huge and complex task of supplying parts for many differe.t brands of many types of equipment; this must work against the important efforts made to mechanize Algerian farms. 7.15 The state agency responsible for the sale and distribution of farm equipment and implements recently licensed a small number of private shops to sell spare parts, do repairs, sell small agricultural implements, and provide guaranteed after sales service, subject to prices and margins set by Government. Unfortunately, consideration is now being given to abolishing the right of private traders to stock spares for resale giving an exclusive monopoly to PMA (the national monopoly manufacturer of agricultural equipment) and the EDIMAs (wilaya level public enterprises set up to sell and service farm equipment). As with other inputs, it is recommended that Government (a) place priority on allocating sufficient foreign exchange to import spare parts, and (b) liberalize the import, purchase, sale and distribution of new equipment and spare parts at margins that encourage private sector participation. It is also ecommended that Government encourage small-scale entrepreneurs to set up hire services, which would lead to better maintenance and higher utilization rates of the farm machinery. This may require changes in current licensing regulations. Finally, it is recommended that Government set prices of spares at a level that does not discourage the repair of old equipment in favor of the a,urchase of new equipment; at present, there is a 40X tariff on import of spares making them relatively more expensive than new equipment. D. Liberalizing the-Marki Ystems 7.16 Until 1982, most agricultural products and foods were marketed by state-owned and operated national marketing bureaus, specializing in one of several commodities. The marketing bureaus, called Offices, include an office for the marketing of cereals and pulses, one for red meat, one for milk and milk products, one for poultry products, one for wine products and table grapes, one for dates, and one for olive and other oil seed products. In addition, a national office exists for the marketing of fresh fruits and vegetables. The national offices, some of which have been recently reorganized into regional offices, operate at the local level through Government-sponsored marketing cooperatives so that a farmer who sells his produce through cooperatives may have to deal with, and even belong to, many different entities. 7.17 It is current Government policy that all cereals, pulses, grapes and olives be sold through official channels at prices fixed by Government. In fact, between 50-70X of local production of ereals andpulseins, covering about 60-70X of the cultivated area, is handled in a 'parallel market" that is  - 47 - officially acknowledged and allowed to exist; these commodities are milled locally for farmers' own or neighbors' consumption. Since local production represents about 25% of total consumption -- the balance of 75% is imported -- the amount of cereals and pulses actually passing outside official channels is quite small, at 10-15% of total consumption. For livestock products, including red meat, poultry products, and milk products, an active private sector exists alongside the official channels. A major transformation took place in 1982 with the marketing of fresh fruits and vegetables. Government liberalized the marketing rf these commodities and private traders have become active in all but certain export crops, such as olives, wine grapes, dates, and citrus fruit. As noted in Chapter III, the liberalization of prices and marketing arrangements of fruits and vegetables has had a positive impact on the production of these commodities. Official marketing agencies were still maintained for fruits and vegetables to buy part of the DAS production; with the breakup of the DAS, their role should become one of providing a floor price in the event of excess supply (see para. 6.17). 7.18 Overall, there has been a clear tendency to liberalize the marketing of commodities. This is likely to be accelerated with the breakup of the DAS, since it is harder to control markets when there are large numbers of small producers than with small numbers of large state farms. It is recommended that Government consider further liberalizing marketing arrangements, especially for barley and other cereals produced for livestock production, most of which already pass outside official channels, and for pulses. Second, it is recommended that marketing arrangements of export crops be further liberalized as has recently been done in Morocco and was already the case in Tunisia for most crops (except for olive oil, which remains controlled)'. Progress has already been made on this front. In the past, all agricultural exports had to pass through state monopolies, while farmers are now apparently free to export directly, although it is not clear to what extent they can retain the foreign exchange (para. 9.18). Finally, it is recommended that Government reduce the role or, at a minimum, rationalize the number and scope of cooperatives and marketing agencies. E. A Mew Role for the Ministry of Agriculture 7.19 Agricultural growth rates in the near term will not be fundamentally affected by the role played by the Ministry of Agriculture, but in the medium and long term the ability of this Ministry to create a favorable environment for growth will be critical for the success of the reforms underway. This In Morocco, Government liberalized exports so that exporters no longer had to pass through a state-owned monopoly. Despite the entry of Spain and Portugal to the EEC, the value of Moroccan agricultural exports has increased, with a shift from citrus to new high-value exports such as flowers and early vegetables. Deregulation has apparently allowed greater flexibility and creativity in findin,g new markets and opportunities in Europe in the face of strong compet:.tion from Spain and Portugal.  - 48 - Ministry, together with the Ministry of Equipment, will have to establish priorities and criteria for investments, to ensure an adequate incentive framework to encourage private investments, and to plan and monitor public investments. In the past, a large part of the work of the Ministry of Agriculture was oriented toward planning, supporting and monitoring the DAS in the production process itself. With the dissolution of the DAS, the role and focus of the Ministry's work will change. First, the Ministry needs to build up and maintain a viable research and extension service as discussed above. Second, it should strengthen its overall planning caRacity to establish sectoral objectives and priorities for development, both for private on-farm developments and for centralized public investments (Chapter VIII). For example, sectoral priorities should be examined, not only in terms of their ability to substitute for imports or maximize production, but in terms of their economic and budgetary implications (Chapter IX). These priorities will then have to be communicated by use of incentives. Third, therefore, Government should create a system of prices and incentives that sends signals to farmers and which maximizes economic returns to the country (Chapter IX). To this end, the Ministry will need to strengthen its analytic capacity. 7.20 Fourth, in the context of a favorable environment for agriculture, the Ministry should allow non-government support services of credit, input supplies, and marketing to function efficiently to provide services to farmers at reasonable cost. The Ministry should review existing marketing arrangements and continue to monitor inpu; supplies through the newly created "real" cooperatives with a view to increasing efficiency and reducing the burden on Government of operating these agencies (paras 7.09-7.15). In the future, the Ministry will have a continuous responsibility to monitor the efficiency of the systems for input delivery and equipment services, identify appropriate consultants and to introduce required reforms in legislation and/or changes in the application of the legislation. To make tne decentralized system work better, the Ministry, together with the Ministry of Equipment, also has a responsibility to monitor the use of inputs for their environmental impact (see para 8.25 for further discussion). 7.21 In accordance with the move towards decentralization, most investments will now be part of the provincial-level plans (see Chapter VIII), with the larger investments still planned at the central level. As a result, the role of the Ministries will change from planning the investments to monitoring them. It in therefore leommended that the Ministry of Agriculture and the Ministry of Equipment build up their review capacity both at the wilaya and central levels so that all nationally funded projects of a certain minimum size can be reviewed against a set of investment criteria, including tachnical, financial, and economic viability, and social and environmental impact. The need to strengthen the economic basis of decisions is particularly urgent in the Ministry of Equipment (para. 8.23). 7.22 Fifth, in addition to introducing criteria for investment planning and decision-making, a public investment/eNeniture review should be undertaken to assess funding commitments for all incomplete projects currently underway and  - 49 - to rank these in order of priority. In February 1990, the joint Government/Bank started work on a review of the public investment program. The recommendations of this mission will be discussed in April or May 1990. 7.23 Finally, the Ministry (and the Ministry of Equipment) has the responsibility for monitoring perfQrmance of the agricultural sector, including statistical data collection and analysis, and environmental monitoring (paras. 8.25 and 8.33-8.35). The DAS provided selected statistical data on area and production. It is not clear to what extent the reorganization of the agricultural sector will weaken data collection. It is strongly recommended that the Ministry strengthen its capacity to monitor the agricultural reform program so that it can identify constraints and take corrective action in a timely manner. In this context, monitoring should also focus on the performance of supporting institutions (for example, to ensure the timely availability of fertilizers). It should also focus on the impact of present and future Government policies. In summary, to ensure long-term growth rates for agriculture, the Ministry needs to concentrate on and strengthen its capacity to: (a) plan, evaluate and monitor the sector to ensure a favorable environment for agriculture and (b) ensure the provision of adequate services and incentives for farmers. This may require strengthening and training staff within the Ministry. F. Livestock-Related Issues 7.24 Government's objective, as it was stated in the 1985-89 Plan, is to expand livestock production and thereby satisfy domestic requirements for poultry, eggs, honey, 70X of the demand for red meat, and 36% for milk and milk products. As noted, growth in the livestock sector has generally been good, especially in the poultry sub-sector and has been an important source of growth for the sector as a whole. This growth has been achieved primarily in the private sector, spurred by high prices, liberalized markets, and availability of credit for investments in egg production. The level of prices for livestock products is a major issue in the livestock sector and is discussed in Chapter IX. In addition, there are a number of technical and institutional issues relating to milk and meat production. 7.25 Production of milk, the principal imported livestock product, has increased slowly: domestic production, mainly from cattle, grew at 2.8% per year between 1967 and 1988. To encourage milk production, emphasis was placed on the importation of exotic cattle for the socialist sector. This policy has been helped by the EEC farm policy which subsidizes cattle in Europe, the over-valued exchange rate making imports of cattle and cattle feed relatively inexpensive (imported cattle cost about 60% that of local cattle), and producer subsidies for milk in Algeria. While it is certainly financially viable at present, the policy of importing large numbers of exotic cattle raises two issues. First, Algeria is not self-sufficient in animal feed. Maintaining these exotic animals requires import of feed and an increase in the area under fodder crops. If Government had a pricing policy that reflected import parity prices, it is not sure that either of these options would be economically viable (Chapter IX). Moreover, except in certain limited highly favorable areas, Algeria may  - 50 - not have a comparative advantage in cattle production, while it probably does for sheep and goat production. Second, exotic animals require much higher standards of nutrition and management than local or cross-bred animals. They are also more susceptible to disease. In brief, a more active program of absorptive crossing with improved bulls or, in special cases, artificial insemination (Al) may be more desirable for Algerian conditions than the import of exotic animals. Farmers should, however, remain free to import exotic animals. If further analysis shows this to be uneconomic for the country, however, Government should not encourage importation of exotic breeds; nor should it provide foreign exchange or credit for the importation. It is recommended that this issue be further examined by the Bank and Government. 7.26 As with other inputs, the DAS received a disproportionate share of livestock inputs and services, much of which was provided through livestock cooperatives. Unfortunately, relatively few private livestock owners are members of these cooperatives. To reach the private sector, it is recommended that Government expand its extension effort to (a) provide technical advice to livestock owners; (b) help with the formation of land user groups in communal grazing areas; and (c) assist in obtaining inputs. In this context, it is recommended that private traders be allowed to play a more active role in the supply of livestock inputs (para. 7.12). Regarding health services, it is recommended that Government expand its capacity to provide preventive public health services through, for example, vaccination campaigns and better supervision of slaughterhouses. With an annual turnout of about 200 veterinarians, Government's requirement for such staff is likely to be fully met within 2-3 years. Government plans to encourage veterinarians to set themselves up in private practice; credit would be available through BADR. It is understood that some veterinarians may have already done so but are constrained by shortage of drugs and other veterinary supplies. It is recommended that Government promote the use of private and cooperative veterinarians by ensuring access to sufficient veterinary supplies. Private veterinary services are being tried on a pilot basis in Tunisia and Morocco with considerable success. Finally, regarding services, both the private sector and cooperatives should be encouraged, possibly along ANAND lines in India, to set up milk collection centers with concomitant input services. A Bank-supported irrigation project in Mitidja is likely to spur private investment in livestock in the area. It is recommended that this be monitored carefully for technical and economic replicability. 7.27 Finally, now that Government is disengaging from direct production- related activities, it is important that the Ministry of Agriculture focus on the environment for livestock development: (a) pricing and taxation policy; (b) strengthening livestock extension and research services; (c) training staff; and (d) availability of credit, which is the major investment tool in the livestock sector.  -51 - VIII. Private and Public Investments: Priorities for the Future A. IntrQduction 8.01 As shown in Chapter IV, increased yields will provide the major source of growth in the short and medium term. The previous two chapters summarized the issues that touch on the reform and the future evolution of the EACs, and the actions that Government will need to take to ensure that farmers have the necessary support services. Price incentives are discussed in Chapter IX. The other major source of growth in agriculture, in addition to increased yields, which would affect agricultural growth rates in the medium term, before 1995, is investment. The investment can take two forms: private investment, financed either through savings or credit, and public investments, usually in larger-scale, infrastructure-related projects that would not be undertaken in the absence of Government intervention. 8.02 In the past, most investments in agriculture were undertaken by the public sector and were aimed largely at the DAS and, since about 1983, at expanding the area under irrigation: most irrigation was on land operated by the DAS. Credit to private farmers was a relatively minor, although growing, source of investment. Government has now changed this orientation, with most investment being privately financed. The allocation of credit for equipment purchase and other small-scale investments is now determined by the banking system and the investors, which places new responsibilities on the nascent banking sector and on the farmers (discussed below). Indeed, aside from the dissolution of the DAS and the increased liberalization of the sector, the most important reform undertaken to date affecting agriculture concerns the banking sector, which, provided that inputs are available, should now become an important engine of growth in agriculture. B. Exnanding Private Investment through Credit 8.03 Until recently, most Algerian banks acted merely as conduits to channel Treasury credits to finance investments and operating losses of state enterprises. This was also the case with BADR, established in 1982 to channel more resources to agriculture in general and to private farmers in particular. BADR's main task became, however, financing and monitoring the financial liquidity of the DAS, the state marketing agencies and the Government cooperatives. Little attention was paid to the creditworthiness of the DAS and public sector agencies, as their loans were implicitly guaranteed by Government. This guarantee was made good in 1987 when the DAS were dissolved and their debts assumed by Government. With the Banking Law of 1986 (and amendments of 1988), the framework was established for BADR to become completely responsible for mobilizing its own resources and for making loans; with a few exceptions for special programs or strategic investments, BADR can no longer rely on Government to cover deficits. In early 1989, BADR became autonomous. 8.04 There are longer-term issues of institution building and financial solvability for the credit-delivery system. A beginning has been made with an agricultural credit project supported by the Bank, which is designed to address  -52 - the development of BADR's capacity to appraise projects, to monitor repayments, to keep detailed accounts, to develop a management information system, and to prepare a medium term plan. For the immediate term, BADR must face the simultaneous challenge of: (a) being transformed from a conduit for Treasury funds to a fully autonomous bank responsible for its loans, profits and losses; (b) dealing with 27,000 newly-created EAC/EAIs with no history as farm owners/managers or as borLowers; and (c) continuing to expand its lending to small-scale private farmers, who, by 1989, numbered over 100,000 as BADR borrowers out of 700,000 - 1,000,000 farmers. As noted in para. 6.07 there has been a strong demand for credit from the EAC/EAIs, for livestock, sinking of tubewells, purchase of farm machinery and equipment, and the establishment of orchards, as well as for short term needs. 8.05 Since the agricultural reform, BADR has met the demand for credit where possible (where the inputs and equipment were available for sale), and the amount of lending both for seasonal and investment needs has increased. The last two years have been difficult for repayment, because of the drought, and BADR has twice allowed a moratorium on repayment (in 1988 and again in 1989). It will be important to monitor closely repayment performance and BADR's financial situation in the next several years. The risks are that if BADR suffers low repayment rates, it will quickly grow overly-cautious and/or become illiquid, which will immobilize the agricultural credit system. It is crucial that Government treat this period as one of transition for BADR as well as for the EAC/EAIs. In spite of the new autonomy of both BADR and the EAC/EAIs, it is recommended that Government be prepared to provide support, in the form of enabling BADR to roll-over debts and/or providing support for the Guarantee Fund, on a declining basis, and only for a transitional period of up to five years. It should be emphasized, however, that this support should not be automatic; it should be based on a case-by-case justification. In the longer term, BADR may need to address the issue of using land as collateral, the absence of which has proved a constraint to the development of properly fu:ictioning credit markets in some other countries; the issue is discussed in Annex 2. 8.06 To create a more favorable environment for private investment, Government has, since 1982, introduced new legislation to clarify the role of the private sector in the Algerian economy. These reforms have included easier access to imports for private sector investors (introduced in late 1984), removal of the ceilings on the volume of investment undertaken by private entrepreneurs and removal of the ceiling on bank credit available to investors. An attempt has also been made to simplify investment approvals and import licensing by allocating this responsibility to the national and regional Chambers of Commerce, although foreign exchange is still rationed, by bureaucratic procedures, by the Chamber of Commerce. There is nevertheless, a new optimism in the private sector outside agriculture, which could affect agriculture to the extent that investments in agro-processing and agroindustries are forthcoming, thereby providing an expanded market for fruits and vegetables. 8.07 Interest Rates. After seven years (1979 to 1986) of fixed interest rates, Government increased them in October 1986 by about 50%. They were again increased by 30-40% in May, 1989, and are now 10-12.X. Loans for the production and storage of cereals and pulses are generally up to 8%. While some rates remain negative in real terms (official inflation has been running at around 10-  -53 - 12% p.a.), the increases since early 1986 are significant. Moreover, the Government is committed, within the next 6-8 months, to establishing a timetable for the gradual reduction of interest rate distortions through selective cred-t. It has also indicated its willingness to increase rates further over the next three years so that all rates become positive in real terms. The structure of interest rates has, in fact, played almost no role in allocating resources in the past, since this was done by central planning on the basis of allocation of physical investments. Resources to and from the private sector have represented less than 20% of total resource flows in the banking system. While setting up an adequate interest rate stcucture is important both for ensuring a viable financial sector and for sending signals to an increasing number of autonomous enterprises, the structure of interest rates and its impact on allocation of resources must be seen in the larger context of the overall pricing framework and the importance of pricing signals for resource allocation. This applies in agriculture as well, where many, but not all, output prices are liberalized, but where input prices are not (Chapter IX). 8.08 In the medium and longer term, commercial banks other than BADR are expected to play an increasing role in financing investments in the agricultural sector. Thus credit delivery will become more competitive, which should benefit farmers. The extent to which it may be possible to expand the volume of credit to the agricultural sector from commercial banks, and to a lesser extent from BADR, depends, in part, on the availability of 'selective credit" for agriculture. This issue is related to the particular nature of lending for agriculture which has high risks and costs. C. The Future Role of Public Investment 8.09 In parallel with the expanded role for private investmert, public investments are decreasing in relative importance. Furthermore, in response to Government policy to strengthen its regional focus, most public investments -- with the exception of major investments such as dams and large scale irrigation -- have been decentralized to the 48 wilayat so that investments respect local priorities and needs. The shift away from centrally-planned public investments represents a major change and reflects Government's long-term strategy for further autonomy of enterprises. 8.10 Public investments in agriculture, defined to include forestry and irrigation, have, in the past, been financed from two sources: - the budget, in the f -m of grants from Government, mainly for non- revenue-generating activities (concours definitifs) and; - credits that were channeled through the banking system and which were part of Treasury oRerations (concours temporaires). These credits, generally for revenue-generating activities, were distinct from funds from the budget. As noted, the decision on which investments would be financed by Treasury credits was made by the Government and not by the banks. Since a large part of these funds was not repaid, these credits were frequently treated like funds from the  -54 - budget -- in effect becoming grants that benefited the DAS. Given the low, or negligible, return to investment in the state farm sector, past public investment in the DAS must be considered non-productive, or highly inefficient. 8.11 Total public investments for the full agricultural sector are shown in Table 14. This shows that planned public investments for the sector financed from the budget have increased by about 35X in current terms in the five years from 1985 to 1990 (the start of the last Plan period and the beginning of the current Plan). In real terms, hogever, this represents a decline of almost 20%. This is a reflection of the decline of almost 40% in the overall size of the budget in real. terms over the same period. In agriculture (excluding irrigation and forestry) the main changes are twofold. First, the decline in the budgetary allocation as Government no longer has to support the DAS and as Government's role changes to one which aims to provide a supporting environment for the sector by building up services, such as research and extension. Second, there is a major change in investments financed by credit. In the past, the amount of investments financed by Treasury operations could be predetermined since investment decisions were made by Government. In future, the now autonomous banks will decide. The amount shown as credit for 1990 is, therefore, only indicative and will depend upon decisions by both farmers and banks. Government has attached increasing importance to the water sector (including irrigation) which now consumes 14% of the total budget against 10% in 1985. It is estimated that about 40X of the budgetary allocation for the water sector is for irrigation. The forestry sector has also slightly increased in importance when measured as a proportion of the total budget. The main change, however, has been in the way forestry investments are planned and implemented; they are now much more decentralized than they were in the past. Table 14: Planned Public rnvestments 1990 (in Hillions of dinars) (in Millions of dinars) Budtet Credit Total Bufdet Credit Total Total Planned Public 48300 57134 105434 50012 63673 113375 Investmenta (all sectors) of which Agriculture 1029 3071 4100 800 3200 4000 (I) (2.1) (5.4) (3.9) (1.6) (5.0) (3.5) Water1 4764 0 4764 7200 100 7300 (x) (9.9) (0) (4.5) (14.4) (0.2) (6.4) Forestry 815 86 900 1000 30 1030 (X) (1.7) (0.1) (0.8) (2.0) (0) (0.9) Total 6608 3157 9765 9000 3330 12330 (X) (13.7) (5.5) (9.3) (18.0) (5.2) (10.9) W/ ater invoetments include those for irrigation and water supplies. In 1990, of the DA 7.2 billion allocated to the sector in the budgot, about DA 2.9 billion are for irrigation (DA 1.7 billion for LSI and DA 1.2 billion for SSI); the balance is for water supply. Information is not available on the amount of investments planned for Irrigation alone in 1985.  -55 - 8.12 As public investment decisions are decentralized, the wilayat are playing a key role in investment planning and decisions; the exceptions to this rule are large-scale investments that remain centralized (paras. 8.13-8.32). This new planning process culminates in the Wilaya Development Plan (WDP), which includes all investments to be implemented, other than those that remain centralized. It represents a new planning tool which enables local priorities to be taken more fully into account. However, it makes the implementation of certain national programs, particularly those in forestry, more difficult to implement and monitor -- an issue that will need to be monitored carefully to see if selected activities should not be recentralized. Within the WDP, investments are financed through: (a) the Wilaya's own budget funded through local taxes and fiscal transfers from the Ministry of Interior; or (b) the "budget decentralise" which is effectively the budget of the various sectoral Ministries, allocated to the wilaya for planning and execution. The WDP is prepared by the Wilaya Executive, approved by the Wilaya Assembly (APW) and then, through an iterative process, is finally agreed with the concerned sectoral Ministries and the Planning Commission. This system of budgeting is new and is still in a state of flux. 8.13 As noted above, large-scale investments will continue to be funded centrally. This includes investments in general studies, research stations, large-scale irrigation, forestry/environment, and certain investments aimed at regional development, especially in the South. The more important of these are discussed below. 8.14 Investments in Irrigation. Relative to Morocco and Tunisia, Algeria lags in development of irrigation. Under present conditions, about 3-400,000 ha are currently irrigated in any year, representing less than 40% of the physical potential, tentatively estimated at about 8-900,000 ha (including the Sahara - para. 8.27). If its full potential is realized, about 11% of Algeria's arable land would be irrigated compared to 19% in Morocco. Although irrigation now covers only 8-10% of the total cultivated area in any year, some 30-40% of current value of agricultural production comes from irrigated land. In addition to its high productivity in the semi-arid Mediterranean environment, irrigation also serves an important social and economic aspect as a "drought-proofing" element in a country where most of the agricultural production is severely affected by erratic rainfall. The impact of irrigation could even be higher if land and water productivity in the currently irrigated areas were at optimum levels, which has not been the case so far. 8.15 There are basically two tvDes of irrigation: the large-scale irrigation schemes (LSI) and the small-scale irrigation schemes (SSIV. Large-scale irrigation currently comprises eight schemes ranging from 3,000 ha to 11,000 ha each, with a total effectively irrigated area of 60,000 ha. Water is regulated by large storage dams and supplied all year round to the farms by modern 'I A third type of irrigation, not discussed in detail here, is spate irrigation, accounting for an additional 100-110,000 ha, caused by occasional floods, in the South Atlas plateaux.  -56 - hydraulic systems. However, most of the LSI schemes were constructed before Independence, and maintenance and operation have become increasingly costly. Until recently, modest budget allocations and low water charge recovery allowed for only minor day-to-day repairs, resulting in frequent shortages of water delivery. At field level, water use has been far from efficient, particularly in the DAS, which occupied most of the LSI schemes. The dissolution of the DAS into small and manageable EACs should prove to be a critical factor in boosting land and water productivity in the LSI schemes. 8.16 About 230,000 ha are covered by small-scale irrigation schemes (SSI) which can be modern or traditional and vary in size from a few to several hundred hectares. Water is supplied from traditional shallow wells (120,000 ha), valley springs (8,000 ha), small rivers through traditional diversion weirs and modern individual pumps (36,000 ha), and recently state-constructed small earth dams (4,000 ha). Also included in SSIs are Saharan oases irrigated by tubewells (60,000 ha). Except for tubewell irrigation, water in SSI is mostly seasonal with virtually little flow in summer. Most of the SSI schemes of more than 100 ha and Saharan tubewells were implemented through public investments before Independence and run by local government since then. They all need rehabilitation. Although reliable data are scarce, the impact of SSI schemes is probably i.mportant as they provide the only source of food and income in the oases, conrribute substantially to the subsistence of rural communities in mountain areas, and produce a significant portion of vegetables and fruit around the large urban centers. 8.17 Potential for expanding SSI is likely to be limited. All existing small rivers and shallow aquifers were developed long ago. Expansion would mostly rely on construction of small earth dams and tubewells: a program for construction of 1,000 small earth dams is underway and few regions offer many additional sites. Past experience also shows that costs are high and sustainability questionable. Most of the potential of the northern aquifers (over 1.7 billion i3) is already exploited and additional groundwater can be found only in limited quantities at increasing depths. Moreover, water supply to rural communities is always given priority over irrigation. In conclusion, pending the results of the ongoing survey undertaken by the Bank-financed Irrigation Engineering Project, it is tentatively estimated that the total SSI area would not exceed 250,000 ha. The development of Sahara irrigation outside the oases is discussed separately below (see paras. 8.27-8.30). 8.18 Irrigation development in Algeria would therefore rely mainly on large-scale irrigation for which physical potential exists for 500,000 ha in total (against 60,000 currently irrigated) if 59 dams were to be constructed for irrigation only and 14 dams were to be constructed for water supply and irrigation. For reasons explained below, it may not be possible or desirable to develop all of this physical potential. About 90X of the irrigated area would be concentrated in the coastal strip where suitable soil and water resources are available. In contrast, there appears to be little scope for irrigation in the arid plateaux. 8.19 Government strategy for irrigation development has substantially evolved since Independence in response to changing priorities. From the early 1960s to around 1980, considerable funds were allocated to water resources development  -57 - with the primary objective of meeting demands for water, urban and industrial growth. Major capital investments were made in water storage transfer and distribution benefiting large cities and industrial areas, while irrigation was given a low priority. During this period, the total irrigated area in Algeria virtually stagnated because the slow development of the newly irrigated areas was offset by the decline in the use of land already equipped by obsolescent systems. This trend was reversed in the early 1980s when Government increased its budget allocation for irrigation and started accelerating rehabilitation work, launching new projects and drawing up physical masterplans. An ambitious SSI program for construction of 1,000 small earth dams was implemented. Results are not yet visible in terms of additional irrigated areas under LSI due to the priority given to rehabilitation of existing systems and the long project cycle from design to construction for any new LSI scheme. However, ongoing programs suggest that an additional 50,000 ha under new LSI schemes could be completed from 1988 to 1995, which would be a remarkable performance. Government has indicated its objective to complete as much as 20,000 ha/year with a view to have the potential 500,000 ha under irrigation by 2010 (excluding the Sahara) but this target is unlikely to be achieved within this timeframe because of implementation and budgetary constraints as discussed below. 8.20 S'gnificant progress has been made by the Government in setting up an adequate institutional framework for irrigation development. The Ministry of Equipment, formerly the Ministry of Hydraulics, was created in 1980, and with specific departments for SSI and LSI. In 1985, Government established independent "Offices des Perimetres Irrigues" (OPI) at the regional level to be responsible for operation of all LSI schemes. It is Government's intention that these Offices be funded solely through the recovery of water charges, a major commitment to resolve this long outstanding problem. To this end, Government increased charges on water for irrigation by 56% in 1985 and by 100X in 1988. While this covers about 60% of the O&M charges, Government has undertaken to recover 100% of O&M costs and part of the capital costs under recent Bank-financed irrigation projects (Cheliff and Mitidja) -- again an area of profound reform. Government als3 established in 1988 a semi-autonomous agency, "l'Agence pour la Gestion de l'Irrigation et du Drainage" (AGID) with its own budget and staff, to remove bureaucratic bottlenecks in the design and implementation of irrigation projects. However, AGID is still facing a serious lack of experienced staff and considerable efforts for staff development and training will be required to ensure AGID's implementation capacity for supervising an ambitious expansion program. 8.21 A number of issues relates to the large and rapid expansion of irrigation. These involve water planning and management, budgetary limitations, implementation capacity, economic viability, environmental implications and institutional requirements. First is the issue of water availability and demand. Total surface water available for all purposes after subtraction for SSI and rural water supply in the upper watersheds is estimated at about 12.4 billion/i3, and an investigation of all potential dam sites suggests that 85 sites exist suitable for dam construction, with 45 of these dams already built or under construction. Given the variability of surface flow in rivers in Algeria, the 85 dams could regulate only some 4.8 billion m3 or about 40% of the available water. The balance could be used only marginally for additional seasonal SSI schemes (small earth dams, in particular) and artificial. recharge of aquifers.  -58 - As for groundwater, the northern aquifers with a potential of 1.7 billion m3 are widely tapped, with the result of over-exploitation and continuous fall of the water table in several areas. Projections for 2010 show that water demand and supply would be in balance nation-wide with 500,000 ha LSI and 45 million population (requiring 3.4 and 1.4 billion i3, respectively), provided no important water sources are lost to pollution (para. 8.25). To a'roid regional imbalances, which already exist in certain areas, a strong water planning and management capacity should be established region-wide to limit further water deficits. In short, competition for water between domestic, irrigation and industrial users is becoming a major issue which will need to be continually reviewed. As all conventional water sources will have been fully developed by 2010, it is recommended that a major effort should begin now to introduce measures to conserve water, by setting realistic water charges (Government has recently increased charges for water for domestic use by 67%) and by encouraging efficient irrigation methods. Furthermore, Government should implement an aggressive approach to recharging the aquifers by using the large unregulated water resources and taking steps to recycle waste water for irrigation. Finally, a long-term plan should be developed for the integrated use of water sources by the various end users -- agricultural, industrial, and domestic. 8.22 Second, as noted in the paragraph above, Algeria has pushed ahead rapidly with its construction of new dams. Unfortunately the planning and implementation of associated infrastructure such as the construction of irrigation systems, has not kept pace. As a result, a growing gap has developed between dam construction and actual utilization of the water either for irrigation or for consumption. While this may give the appearance that investment in, say, irrigation infrastructure is economic (because dams can then be considered as sunk costs), a project that takes into account all costs may not be economic. Indeed, it will almost certainly not be economic when full account is taken of the resulting delay in benefits. It is recommended, therefore, that the Ministry (a) put priority on the completion of existing projects -- those in which dams have already been completed or where dam construction has started - - by building the associated irrigation or water supply infrastructure; (b) delay the construction of new dams if necessary to achieve this; and (c) take steps to strengthen its planning capacity to avoid the long gaps between the completion of dams and the completion of associated infrastructure so that benefits can be brought forward. 8.23 Third, the cost of developing LSI is high, on the order of $12,000/ha (excluding costs of dams). Small earth dams and tubewells for SSI are also expensive. Given current yields in irrigation schemes in Algeria, irrigation projects may have a low economic profitability. Fourth, the high investment cost of irrigation would have s.gnificant budgetary consequences. Government intends to expand the area under LSI at about 20,000 ha/year. Given a base cost of $12,000/ha, this implies an annual investment cost of $240 million excluding the cost of dam construction, related infrastructure and on-farm development. Given the past rate of implementation (about 5000 ha per year), it not certain that this objective can be achieved (para. 8.24). Nor is it certain that such a high level of investment in irrigation could be funded given that this would be much higher than recent allocations or that a sufficient number of economically viable projects can be identified to maintain the planned 20,000 ha/yr. program. Government has drawn up physical (and regional) master plans for water use which  -59 - need to be costed and reviewed in the context of their economic benefits and costs and budgetary consequences. It is recommended that these plans be refined so that they incorporate a 10-year time-horizon for irrigation development that would clarify strategic options, define irrigation priorities in light of their economic costs and benefits including the development of a sound investment program, and review the potential application of different technologies. Government has agreed to prepare such a plan as part of Lhe Irrigation Engineering Loan (2978-ALG). Assistance will, however, be required to undertake the necessary economic analysis, including collection of data. Given the high cost of irrigation development, it is strongly recommended that (a) the Ministry of Equipment (ME) strengthen its capacity to assess the economic viability of investments in the irrigation sector; (b) ME seek assistance, in the short term, to assess the relative economic priority of projects; (c) high return projects be selected first; and (d) attention be given to finding lower cost solutions than have been used in the past. 8.24 Fifth, implementation is hampered by lack of coordination, shortage of trained staff, and shortage of essential materials required in the construction of LSI. As noted above, as a result of implementation delays, a growing gap has developed between completed dams and the starting of construction of associated infrastructure to bring water in these dams into use. Coordination between departments involved in the planning and execution of schemes needs to be improved -- an issue related to the shortage of trained staff. To this end, it is recommended that the Ministry undertake a review to estimate the requirement of trained staff by specialization and match this with available staff. A detailed training program should then be drawn up to upgrade staff skills where necessary, as should a detailed recruitment program. Grant funding might then be sought for such a program. In addition to manpower shortages, the lack of concrete and asbestos pipes for irrigation has been a major constraint to the completion of irrigation schemes -- shortages linked to Algeria's industrial policy. While appropriate steps should be taken to remedy this problem in the medium term, it is recommended that the Government adopt a more flexible approach to the import of pipes to overcome short-term shortages; irrigation authorities should be allowed to import pipes when needed to complete a scheme -- the opportunity cost of delayed benefits would certainly make it an economic solution. 8.25 Sixth, the sustainability of water development is at risk b.3cause major water sources may be lost to pollution or to reservoir sedimentation. In addition, irrigation development involves a number of environmental issues which are exacerbated by the fact that most of LSI development will take place in the narrow and over-populated coastal area. Irrigation brings with it considerable intensification of agriculture which, in turn, leads to: greater use of fertilizers and pesticides which can pollute rivers and groundwater, increased risk of water logging in the higher rainfall areas, and increased risk of salinization of soils in low rainfall areas. As discussed in a recent Bank report on Environmental Problems and Issues in Algeria, there has already been a considerable increase in pollution of rivers with industries and urban areas frequently pumping waste directly into rivers. Runoff from agricultural areas has added to pollution; monitoring of groundwater shows heavy buildup of nitrates in water used for drinking. Pollution must be controlled to avoid loss of drinking water supplies and to avoid the possibility that some surface water  -60 - may become unsuitable for irrigation. Pollution can be significantly reduced by use of appropriate fertilizers and pesticides (para. 8.33) and the application of appropriate integrated pest and fertilizer management strategies. Apart from pollution to rivers caused by runoff from agriculcural areas, the danger also exists of pollution from the manufacture and disposal of fertilizer and pesticides. It is recommended that Algeria's capacity to wonitor surface and groundwater quality be expanded. This will need to be linked to institutional, legal and other measures to control pollution -- issues addressed in the recent report on Environmental Problems and Issues. Siltation of dams through erosion of watersheds due to overgrazing, deforestation and overcultivation is also a concern that can only be reduced by: proper land and water resource management (para. 8.34), a strong extension service that can deliver messages to farmers about soil conservation and means to reduce erosion, and a strong forestry service to help protect upper-catchment areas. 8.26 To deal with these issues it is recommended that the planning capacity at the central level be strengthened to coordinate, in particular: (a) water supply and irrigation projects; (b) use of groundwater and surface water; (c) infrastructure construction and agricultural development in LSI schemes; (d) economic evaluation and setting priorities for irrigation projects; and (e) environment protection and water development. To this end, the Irrigation Engineering Project includes the preparation by ME of a 10-year irrigation plan which would provide a framework to consider all these issues. It is also recommended that regional capacity in water planning and management be developed to refine regional land use and water development master plans and to set up the appropriate mechanism for regional integrated water management; and that it constantly review the environmental implication of irrigation in each region. 8.27 Development in the South. For social reasons and to stem the flow to the already densely populated northern strip of Algeria, Government has recently given increasing emphasis to development of the South (the area south of the Atlas mountains comprising largely the desert and oases areas). The population of the south is low with 7-8% of the total, and as shown in the estimates in Chapter IV on the contribution of growth rates, the impact of development will be largely self-contained, except for date production, which may be exported. The 1985-89 Plan focused on the agronomic development of the region, particularly through land distribution, replanting existing date plantations, and drilling deep wells to experiment with large-scale cereals production. The first of these programs, Accession a la Propriete Fonciere (APF), is aimed at expanding the occupation and utilization of the South and center-south by granting land titles to farmers to develop land in these areas. As of end-1989, about 200,300 ha had been distributed under this scheme. The second program involves public investment, and is essentially an irrigation development program, with extension, research and monitoring aspects to be given some priority. This program is still in the planning phase. The third scheme has involved experimental development of large-scale farms. The intention was to develop 70,000 ha under this five- year plan, and to date about 10,000 ha have been developed through sinking deep wells. A number of major issues are, however, associated with all of these schemes, which affect development of the entire south.  -61 - 8.28 The first issue relates to exploitation of the existing fossil-based groundwater resources, which are not renewable. While water resources are estimated to be large, especially in the far south where conditions for development are particularly harsh with extremes of temperature, their exploitation needs to be prudently and properly controlled. A study (ERESS), updated in 1984, establishes quantitative limitations by sub-zone on water use that would prevent deterioration of the water resources and avoid increasing operating costs. Development of the area already distributed under the APF program would involve the mining of water which would exceed by far (up to four times) the estimated potential resources in the particular sub-zones. While water could be transferred from the southern part of the South, this would add to the expense. In addition, development of the entire area may affect use of the water table by Tunisia. This highlights the critical need to have a comprehensive plan of development which preserves the water table and which is strictly adhered to. It is strongly recommended that before any further exploitation of water resources in the South is undertaken, such a plan be established and the necessary monitoring systems be put into place. 8.29 The second issue is that development of the South is costly. The region is vast, has limited infrastructure which would have to be built up, and lacks skilled labor. This is another reason for a development plan that establishes priorities for development (potable water over irrigation water; rehabilitation over expansion, etc.) and establishes economic criteria for investments, taking into account social and regional distribution aspects. 8.30 Third, the region is ecologically fragile, the soil is short in organic matter, and developments need to be protected from sand intrusion, especially from hot sunmmer winds. Again, a comprehensive land and water resource management plan for the South should take into account these environmental aspects. Soil studies need to be conducted prior to land development to identify tracts of land whose development will not result in the excessive leaching of irrigation water into the groundwater table. Water application rates and techniques need to be determined by applied research to avoid salinization. Since groundwater resources play a fundamental and almost unique role in development in the region, the plan should focus on the special characteristics of these resources: to what extent they may be renewable, their quality, and regional and international implications of development. In addition, and as noted in para. 8.25, it is recommended that the monitoring and control of groundwater resources be strengthened. Government has approached the Bank to support specific developments in the South. This would provide a vehicle to deal with these crucial issues and to ensure that the preparation of a development plan for the South precedes or is parallel to any investment operation. 8.31 Forestry. A third aspect of public investment which would affect the agricultural sector is development and improvement of forest resources to, inter alia, protect the environment, improve vegetative cover particularly in the upper catchment areas, and increase productivity. Forestry products and environmental protection have not been taken into account in estimating the growth rates in Chapter IV. Nevertheless, investments in forestry and in land improvements in  -62 - upper catchment areas would increase infiltration rates thereby adding to groundwater availability, reduce soil erosion and damage to irrigation systems, and increase production of wood products as import substitutes. 8.32 The objectives of public investments in forestry activities would be to: (a) protect existing forests; (b) reduce runoff and thus increase recharge while protecting watershed areas from erosion; and (c) improve the management and exploitation of industrial forests. With respect to the first objective, the main threat to the existing forests is forest fire; because butane gas is readily available at affordable prices, over-use of forests for fuelwood and charcoal is not a problem in Algeria. In recent years, Algeria has made progress in reducing forest fires, but additional efforts are needed. Second, in order to reduce runoff and manage watersheds more efficiently, a study should be carried out of the needs and priorities to identify the major areas for erosion control, and to develop plans to provide protection. This should include an assessment of the potential impact of decreased fallow and possible overgrazing in selected areas. It might also include a pilot land and water resource management project to assess means to better use these resources in a way that can be replicated in other areas of the country; such a project could be financed by the Bank as noted below. Third, to exploit fully the potential of tl,e industrial forests, it is necessary to review and possibly update the existing data and the planning framework for forestry exploitation. It appears that productivity of Algerian forests can be increased, by reducing the costs of exploitation and increasing the yields. The pricing of forestry products needs to be examined, along with the financial situation of INTF, the Government agency responsible for exploiting the forests. Finally, the quantity and value of imports should be examined to determine the scope for import substitution. At present, Algeria imports about US$2-300 million in wood products each year. The World Bank is likely to preappraise a project in May 1990, that would be a first intervention in the forestry sub-sector and which would address the issues listed above; emphasis will be placed on land use management. 8.33 Environmental Concerns and Sustainable Agricultural Growth. Throughout the discussion of the potential for growth in agriculture and the various sources of growth is a constant theme of environmental aspects of this development. It is important to recognize that more intensive use of fertilizers and pesticides that increase productivity and will be an important source of growth can lead to long term hazards. Current agricultural practices need Lo be improved to protect surface and groundwater quality, workers, and consumers. Both research and extension programs need to emphasize the selection and application of appropriate agricultural chemicals, and the development and adoption of integrated pest management techniques should be given priority in order to reduce the cost of inputs, the risk of water contamination and health risks to workers. Agricultural exports to the EEC will require particular pesticide residue monitoring to assure that shipments will not be rejected for failure to comply with EEC standards. In addition, intensified red meat and poultry production may involve greater use of antibiotics and other veterinary drugs, whose safe disposal is of concern. The World Bank has been working with the UNDP on a first phase study on the management of pesticides (which could be expanded to include veterinary drugs) in Algeria, Tunisia and Morocco to address these important environmental concerns. A follow up to this study as well as the study on  -63 - fertilizers (para. 7.12) should make specific recommendations on types of acceptable and unacceptable pesticides and fertilizers, on their use and proper disposal. 8.34 The various investment programs mentioned each involve a number of environmental concerns on the competing uses of ground and surface water, the preservation of both soil and water quality, and the need to do comprehensive planning and careful monitoring to avoid overexploitation or degradation of the current soil and water resources. To ensure sustainability of Algerian agriculture over the long term, it is recommended that Government train and adequately finance the environmental unit to promote and monitor environment and natural resources management'. In addition, the Ministry of Agriculture should train staff in a number of departments (planning; research, extension and training; agricultural development) in environmental aspects of agricultural activities, and should develop a close liaison with the environmental unit to develop acceptable packages, techniques and approaches to agricultural development that eliminate or rinimize any negative environmental impact. 8.35 This report has identified numerous ways to increase agricultural growth. It has also emphasized that this growth must be sustainable. To this end, management of Algeria's land and water resources should be handled in a comprehensive, integrated manner. As an example, intensification of irrigation needs to be examined in the context of increased pesticide and fertilizer use. Certain investments can also aid in the recharge of aquifers, for example, reforestation, better pasture management, improved soil conservation and cropping practices, and the construction of small engineering works to reduce runoff. In short, growth of agriculture must be balanced so that its viability is ensured in the long-run. It is, therefore, essential that existing agricultural operations and new investments are assessed within the context of carefully prepared land and water resource management plans that examine their impact on, inter alia, groundwater recharge, pollution, erosion and long term sustainability. Conclusions on the Scoge for Public Investments 8.36 As noted above (para. 8.02), in the past most investments in agriculture were undertaken by the public sector and were aimed primarily at the DAS. For the future, Government intends to change this orientation, with most investments being undertaken by the private sector. Establishment of a sound financial policy and the building up of viable credit institutions such as BADR must, therefore, be a priority for the Government. Government, however, still has a crucial role in undertaking public investment that would not otherwise be undertaken by private individuals and which are economically and socially justified. Projects in these areas may be suitable for financing by international lending agencies such as the World Bank. First and foremost amongst these, given its potential short- and medium-term impact, is the creation of an extension and research service; this includes extension and animal health services for livestock. Most other investments have a longer payoff period or '/ Given the current reorganization of Ministeries, it is not yet clear in which Ministry this existing unit will be located.  -64 - the benefits are less obvious. Investment in large and medium scale irrigation is crucial. As noted, investment costs are high and it is imperative that all investments yield an adequate rate of return. Medium-scale irrigation, although less expensive than LSI, often generates benefits that are low in relation to costs; again, all project proposals need to be vetted carefully. Generally, it is to be expected that investments in small-scale irrigation would be undertaken by the private sector. Priority also should be given to understanding the ecologically fragile south before further large-scale investments are made in that region. Following appropriate pilot projects and studies where necessary, investmeni.s to preserve the environment should also be considered high on the priority list. These include investments in forestry and land resource management. 8.37 The above is far from an exhaustive list of potential areas for public investment. There may be scope for investment in crop storage, processing plants, including dairy plants, and smaller-scale investments in infrastructure, research, etc. at the local level, although it is expected that much of this would be by the private sector. Substantial investment also needs to be undertaken in upgrading the stock of human capital and in institutional strengthening. Finally, the Government must give priority to the provision of an adequate quantity of farm inputs: seed, fertilizer, farm machinery, spares, and pesticides -- all essentially recurrent inputs for purchase by private farmers. Nevertheless, in addition to providing the "right" policy environment to ensure that these inputs can flow to farmers (pricing policy, marketing policy, and role of the private sector, etc.), the Government needs to make an adequate provision of foreign exchange. IX. Introducing an Effective Price and Incentive Framework A. Introduction 9.01 If the agricultural sector is to achieve the higher growth rates indicated in Chapter IV, it is imperative that Government has in place a system of prices and incentives that encourages farmers to increase productivity and production. Over the past fifteen years, the terms of trade have generally favored Algeria's agricultural sector. The general level of prices (as measured by the GDP deflator) increased by 55X between 1980 and 1988, while agricultural output prices increased by at least 150X over the same period. For those crops which are still controlled, producer prices are higher than import parity prices, even using a shadow exchange rate (para 9.10 and Table 16). From the late 1960s to about 1974, however, producer prices for crops were more or less static (Statistical Appendix, Table A.41); both prices and marketing of crops were controlled by Government. Since then, prices of controlled commodities have increased significantly while, most importantly, in 1982, Government liberalized prices of fruits and vegetables. As a result, these prices, along with fresh milk and meat, are now largely established on the free market. The generally favorable treatment received by the agricultural sector had, however, for reasons explained earlier, little impact on production of the DAS which were largely immune to price incentives. Given the transformation of the means of production, this is no longer the case.  -65 - B. Crop Prices 9.02 Crop prices in Algeria are high in both financial and economic terms: prices of wheat and barley, the most important crops in terms of area, increased by about 300% between 1974/76 and 1988; the increase is around 500% if the 1989 price is included. Between 1974 and 1980, however, these prices remained roughly constant in real terms; between 1980 and 1989, the increase has been about 200% or about 70% in real terms. As shown in Table 15 below, and in more detail in the Statistical Appendix, Tables A.41-A.42, the controlled prices of vegetables and fruits were increased at much the same rate as those of wheat and barley until 1980. Since then, however, the liberalized fruit and vegetable prices increased even faster than those of the controlled cereals and pulses although this may have changed in 1989 given tb1 enormous increase in cereal prices (paragraph 9.11); the increase in prices of fruits and vegetables was also due to the quantitative restrictions on imports. Generally, therefore, there has been a marked shift in the terms of trade in favor of ag-iculture. Table 15: NOMINAL PRICE INDICES FOR AGRICULTURAL INPUTS AND OUTPUTS 1974/76 1980 1983 1987 1988 1989 averagO CroDs Durum Wheat 52 100 112 216 216 320 Bread Wheat 51 100 113 191 191 286 Barley 51 100 100 212 212 286 Potatoes 56 100 173 254 264 Tcmatoes 60 100 375 475 500 Livestock Beef 48 100 160 300 307 Lamb 45 100 153 282 306 Milk 56 100 125 200 210 Inouts Pertilizers (Am. Nitrate) 100 100 100 241 265 Tractor 65hp 100 100 166 330 330 330 Labor 62 100 127 452 452 GDP deflator 52 100 127 147 155 176 Weighted agric. price index 53 100 148 247 9.03 Until the dissolution of the DAS, it was Government policy to set producer prices of controlled commodities so as to cover the costs of production of the better DAS. This had two implications. First, it meant that the majority of DAS, less efficient than those selected, necessarily made a loss. Second, since yields on DAS were generally low by international standards (at 650-900 kg/ha for wheat, among the lowest in the world relative to major cereals producers, neighboring countries, and countries with similar per capita incomes -- Statistical Appendix, Table A.32), prices had to be high by international standards to cover the costs of production of even the better DAS. Government's future pricing policy for cereals and other agricultural commodities is discussed in paras. 9.12-9.16.  - 66 - C. InRut Pricing 9.04 The primary objective of Government's agricultural input pricing policy was, until 1983, to limit the increase in consumer prices. To this end, prices of agricultural inputs, fixed by Government, were unchanged from 1974 to 1983 (Table 15). As a result, the real prices of fertilizers, plant protection chemicals and equipment declined in real terms by 50X between 1973 and 1980. Since production costs of farm chemicals and equipment rose considerably during the period, Government incurred sizable subsidy costs. In the case of farm machinery and equipment, purchase prices of new equipment were in many cases less than the prices of spare part:s, which were increased annually in line with production costs, thereby encouragtng purchase of new equipment rather than maintenance and repair of existing equipment. It also encouraged the diversion of machines to non-agricultural users (e.g. construction) for which no comparable subsidies were available. 9.05 Government adopted in 1983 a new approach to input pricing: "vdrit. des prix", aimed to bring prices closer to "production costs". Since 1983, prices of inputs have been increased substantially. It was Government's intention to eliminate all subsidies on inputs by no later than the end of 1989. At the date that this report was being finalized (end 1989), data were not available to verify whether or not the object'ves had been met. It is possible to state, however, that starting from fixed ex-factory prices, most subsidies have already been eliminated. Given the continuous adjustment of the exchange rate, input prices will need to be increased on a continuous basis, however. In fact, while all "direct subsidies" have likely been removed, Government defines the policy of "verit6 des prix" to cover ex-factory costs and all marketing and handling costs of inputs. I. is not clear, however, the extent to which the ex-factory price includes direct or indirect subsidies because of protection or subsidies on selected raw materials. Nor is it evident that all subsidies should be removed since the farmers would then pay for inefficiencies of local industry. This issue can best be addressed by linking prices to the world market as discussed below. D. Livestock Pricing 9.06 The level of prices for livestock products is the major issue in the livestock sector: meat prices are three to four times the world market price at the existing exchange rate; producer milk prices are also two to three times the economic price (obviously lower than this at shadow exchange rates; see paras 9.09-9.11 and Table 16 below). Meat prices remain at these levels because of high demand, restrictions on imports, and inefficiencies in marketing arrangements. Consumption of red and white meat combined rose from 8.3 kg per person in 1973 to over 21 kg in 1988. The limited quantities of imported meat are sold in controlled markets at prices that cover costs of importation but which, because of the over-valued exchange rate, are low in relation to world market prices. 9.07 Most milk is imported as powder, reconstituted, and sold at a retail price of DA 1.7/1, which has remained unchanged for the past 10 years. Fresh milk is sold by local traders at DA 5-7/1 because consumers are willing to pay a premium  -67 - for fresh milk even though it is unprocessed. Government also purchases fresh milk through ORIAC, the Office du Lait, from local producers at DA 4/1. Fresh milk purchased by ORLAC is then mixed with the imported milk powder and fats to make reconstituted milk containing about lOX fresh milk. Costs of producing this milk are about DA 2 per liter as of mid-1988 and are sold ex-factory at DA 1.35, for a per liter loss of DA .65. Losses have been somewhat lower per liter in the past. They are financed partly through profits on s&le of cheese and other luxury dairy items and partly through subsidies from the Fonds National de Compensation. GoverTnent apparently intends to increase the producer price of fresh milk further to DA 4.50/1, which would raise the cost of reconstituted milk accordingly. ORIAC has proposed to raise the ex-factory and hence consumer price of reconstituted milk but no decision has yet been made. Pricing is further discussed in paras 9.19 to 9.25. 9.08 While the high level of producer prices has encouraged development of the livestock sector, the Algerian livestock industry may be based on prices that cannot be sustained in the long-run. In addition, an FAO report suggests that the relatively high prices of livestock products have encouraged farmers to grow low-yielding cereals (particularly wheat) for the straw and stubble value for livestock. If internal prices adjust to world market levels, much of the 'modern" livestock industry could collapse, that is, the livestock sub-sectors that rely heavily on imported feed and concentrates -- mainly exotic cattle and some fattening programs. This is probably not true of the poultry sector because of their high value added (they are much more efficient feed converters than ruminants such as beef, sheep and goats). Preliminary estimates suggest that Algeria probably does not have a comparative advantage in cattle production for many of the production systems currently being used although these estimates need to be verified. If analysis should conclude that it does not have a comparative advantage, the national policy towards the encouragement of all types of cattle production should be reassessed. Government policy should most likely concentrate efforts on increasing meat production from sheep and goats. Such a policy would also help to reduce beef prices. It is recommended that Government: (a) liberalize the import of meat to prevent the development of an uneconomic industry behind trade barriers, which would have high social costs should the industry be subjected to world trade in the future; the meat industry could, however, be provided limiced protection through the introduction of an import tariff based on the real (unsubsidized) cost of producing beef on the world market; and (b) freeze the controlled producer price of milk in nominal terms unti' rale price is more in line with world market prices (based on . "shadow' exa_ange rate). E. Comparison with World Market Prices 9.09 As noted, producer prices, both for controlled commodities and for commodities sold on the free market, are high in relation to world prices when prices are converted at the official exchange rate. The comparison with international prices translated into border price equivalents is most easily done for crops traded on the world market: the wheats, barley, pulses and potatoes. I fresh fruits and vegetables, international prices are a function primarily of quality and season and so are not easily comparable. Table 16 below compares domestic producer prices with border price equivelents based on the  -68 - official exchange rate, currently about US$1 - DA 7.5 (OER). The final column gives the border price equivalent based on a shadow exchange rate (SER) of US$1DA 12.0. 9.10 In 1974, producer prices had hardly been increased for seven years and were relatively low. Table 16 shows that in 1974 the economic farmgate prices for crops were considerably higher than the producer prices (nominal protection co-efficients between 0.40 and 0.65). Shadow pricing would have exaggerated these differences. In the mid-1970s, therefore, it could be argued that Algerian farmers were taxed vis-a-vis the world market and that this partly explains the poor performance of agriculture in these years. Since 1980, however, the trend has reversed so that the NPCs are considerably greater than one showing that domestic production of these crops frequently enjoys a high level of protection. The differences are less marked using the shadow exchange rate but are still substantial. The economic prices for cereals and for fertilizer are based on projected 1995 prices (in constant 1989 terms) rather than actual 1988/89 prices to minimize the impact of short term fluctuations in world prices. The average world market price for wheat, for example, was US$161 per ton in 1986, US$134 in 1987, and US$180 in 1988 following the drought in North America, demonstrating considerable year-to-year fluctuation. The projected 1995 price (in constant 1989 terms) is US$146 per ton. If the actual 1988 price (which is significantly higher than the projected 1995 price) had been used in estimating the economic price, the economic price so calculated would have been higher, suggesting a lower level of nominal protection. Table 16 shows that hard wheat is presently heavily protected based both on the OER and on the SER of US$1 DA 12.0. In addition, use of the SER shows heavily subsidized inputs. Table 16: FARMGATE PRICES (DA/TON) 1974 1989 5/ Financial Economic Financial Economic Economic at OER at SER Durum wheat 540 949 4,000 1,690goI 2,53 0V? Bread wheat 480 825 3,300 1 570V 2 350J Barley 317 781 2,300 1, 120V 1 630& Potatoes 5/ 550 821 3,300 1,340 2,000 Tomatoes 5/ 4,000 2,800 3,400 Beef 5/ 35-46,000 12,6008 20 000oI Milk 5/ 4,200 1,850 2 680V Fertilizers 5/ (Am. Nitrate) 924 1,550V 2,260V Tractors (50-70 hp) 82,400 90,000 160,000 1/ With 10X premium for durum wheat 2/ Based on World Bank's projected price for 1995 (in constant 1989 terms); details are in para 9.10 3/ Includes 20S premium for fresh milk and beef 4/ 1987 price 5/ Prices below are for 1989 with the exception of potatoes, tomatoes, beqf, milk, and fertilizer where 1988 prices have been used. 9.11 The considerable short-term fluctuations in world market prices -- the world market price of durum (hard) wheat increased by 35X between April and June 1988 -- explain, in part, Government's desire to reduce their dependence on imports of strategic crops. This is an important objective in the agricultural  -69 - sector and explains the Council of Ministers' decision, announced in July 1988, to introduce a system of payment of premiums for cereals and pulses -- the strategic crops -- to increase production and thereby reduce dependence on imports. In this context, the price of durum (hard) wheat was increased to DA 4000/ton, made up of a base price of DA 3,200/ton (against a 1988 price of DA 2,700) and a premium of DA 800/ton. This was not accompanied by a commensurate increase in consumer prices of milled flour products, implying a considerable increase in the already substantial level of subsidy (paras 9.19-9.25 and Working Paper 2). F. The Need to Rethink Future Agricultural Pricing Policy 9.12 Setting producer prices depends on the objectives of pricing policy. On the one hand, high prices tend to maximize output and reduce dependence on imports. On the other, low producer prices can assist in a Government objective of low cost food for urban populations. It is recommended that when prices are fixed by the Government, which in the medium-term is expected to be the case for only a few crops considered strategic by Government, prices be set at levels that reflect the economic costs to the country of importing or exporting these commodities'. The same policy is recommended for setting prices of those inputs whose prices are fixed by Government, since the subsidization of inputs distorts production away from products with the highest economic return and results in over-utilization of inputs. In brief, controlled input and output prices should be set in relation to world market prices rather than ':o the costs of production in Algeria. Based on the exchange rate used and the degree of protection to be provided to the sector (para. 9.14), such a pricing policy may result in slightly reduced farm profitability in some cases -- depending upon the particular crop or livestock activity o" each farm. At present, farming in Algeria is potentially highly profitabie as evidenced by the number of civil servants who left Government service to join the EAC/EAIs. A slight reduction in profitability is unlikely, therefore, to have a significantly negative impact on production. In the same vein, agriculture is likely to remain relatively profitable even with a modest worsening of agricultural terms of trade brought about by an adjustment of the prices including the exchange rate (Section I below) or a lifting of quantitative restrictions. 9.13 Certain internationally traded commodities are subsidized or dumped as part of farm policies of exporting countries. The EEC, for example, and the United States have support programs for milk powder and wheat which result in traded prices below the costs of production in the most efficient exporting countries. In such cases, prices should be set in relation to long run international producer prices following the example of Tunisia which has adopted the policy of setting wheat prices in relation to a three year moving average of spot and futures wheat prices on the Chicago Board of Trade. A 15X tariff '/ In the past, prices were published, by Presidential decree in the Journal Officiel, quite late in the crop season. It is also recommended that prices of controlled commodities be announced 2-3 months before the planting season so that prices can affect farmer behaviour.  -70 - is then added, which aims to establish an effective protection rate of 25X. In this way, local producers can be protected from dumping and benefit from undistorted world price levels. 9.14 Given the over-valued exchange rate, the presence of quotas/restrictions on imports, and selected subsidies in the industrial sector, it is clear that Government provides a considerable degree of protection to industry. The issue is whether agricultural prices, if set in relation to world market prices, should be set at world market equivalents or whether there should be an element of protection to the agricultural sector as in Tunisia. This would provide a measure of protection to the EACs -- infant industry argument. This issue is also linked to: (a) the rate used to convert international prices to domestic prices; (b) long run expectations; and (c) Government objectives for the sector vis-a-vis other sectors of the economy, employment, farm productivity, rural versus urban growth, and the desire to reduce budgetary outlays and subsidies. Given that, for example, 75X of cereals are imported and half the balance is consumed on-farm, only 10-15X of Algeria's cereal needs are directly affected by incentives. The impact of highly subsidized producer or input prices is therefore relatively small in relation to total needs. It might, therefore, be argued that cereal prices should not be protected, thereby minimizing Government budgetary outlays. If, on the other hand, Government places high priority on rural employment and incomes to reduce urban migration with its associated infrastructure costs (Government would have to provide housing, water, sanitation, etc. in the urban areas), then the "cost" of maintaining high producer prices may be perceived as well worth paying. 9.15 To gain a better understanding and to develop a mechanism for setting prices and incentives, Government has agreed to undertake a study of prices and incentives along the lines of studies recently completed in Morocco and Tunisia. These studies, which have estimated DRCs for a number of commodities grown in different ecological zones, have shown that Morocco and Tunisia have a comparative advantage in the production of fruits and vegetables, including citrus. In Tunisia, 60-65X of cereal production has DRCs of less than one; production was uneconomic only in more marginal areas. Livestock, especially beef production, was also less attractive, as was production of certain commercial crops such as sugar. Because of the very different structure of Algeria's economy which, in the past, incorporated centralized planning, high investment in heavy industry, relatively high levels of mechanization and high construction and wage costs, it does not necessarily follow that what applies for Morocco and Tunisia will apply to Algeria. However, prima facie, it is likely that Algeria has a comparative advantage in fruits and vegetables, some cereals, and sheep production. 9.16 The purpose of the proposed study on Rrices and incentives is to analyze the existing system of pricing and incentives in Algerian agriculture, and examine policy options aimed to increase efficiency of resource allocation. The study should lead to a series of recommendations on, inter alia: - methodologies for establishing agricultural prices; - the comparative economic advantage of producing various agricultural commodities in Algeria; and  -71 - - the consequences and impact of different policy options, including the impact of different rates of protection. Timing and financial arrangements for the study ha'-e still to be agreed. On the innut side, several studies are already underway, the most crucial of which is the study on the supply and distribution of fertilizers, financed by UNDP with the Bank as executing agency (para. 7.12). This study will, inter alia, make specific recommendation on (a) the structure of fertilizer prices, including margins from point of import or manufacture to distribution at farm level and (b) price subsidy levels necessary to meet Government sub-sectoral objectives within a global objective of improving efficiency in the allocation and use of resources. G. Increasing Incentives for Agricultural ExRorts 9.17 One of Government's stated objectives is to expand and diversify non-hydrocarbon exports. Agricultural exports, however, now account for less than 1% of the total value of merchandise exports from Algeria and have been in almost continual decline for the last fifteen years. With adequate price incentives for export crops -- largely a function of the exchange rate -- and the increases in fruit and vegetable production from growth described in Chapter IV, exports can be increased. Increased domestic demand for export crops, particularly dates, citrus fruit, and wine may, however, militate against large increases in exports. In addition, with the recent expansion of the EEC, the difficulties of exporting to Europe, Algeria's biggest market (more specifically France), and the increased competition from Eastern Europe following the recent liberalization of their regimes and from other Mediterranean countries with similar products to sell, Algeria faces a difficult task. 9.18 Although the contribution of" sriculture exports to the balance of trade is small, so that major efforts tf: i-rease these exports are unlikely to have a noticepablc impact on the balance of trade in the near to medium term, efforts should be begun with a view to the longer term benefits, beyond the year 2000. First, it is recommended that an analysis be done of Algeria's comparative advantage for production of various export crops. Second, the incentive structure should be examined: the level of producer prices relative to export parity prices; domestic marketing arrangements by Government-sponsored national agencies; trading arrangements and the ability of the marketing agencies to identify and pursue export marketing opportunities; and the retention of foreign exchange earnings from exports. As noted in para 7.18, Government has liberalized the export of agricultural commodities so that exporters no longer have to pass through state monopolies. Regarding the retention of foreign exchange earnings, it seems that it is linked to the use of foreign exchange in the production of the exports; from industrial exports, for example, exporters can apparently only retain 1OX of the net foreign exchange surplus earned during the previous year. It is recommended that exceptions to this policy be made where strong potential exists for agricultural exports, so that exporters of agricultural products can retain a higher percentage of the foreign exchange earned. The issues of profitability and incentives are closely linked, of course, to the prices at which exporters can sell their products, which in turn depends on the exchange rate.  -72 - H. Consumer and Producer Prices and Subsidies 9.19 While Government has liberalized prices of fruits and vegetables, consumer prices of most "essential" food items such as bread and other wheat products and reconstituted milk are controlled. Government has a policy of providing food to consumers at subsidized prices. In pursuit of this policy, consumer prices were increased only slowly through the 1970s. For example, the cost of bread and milk decreased significantly in real terms between 1974 and 1987. More recently, Government has attempted to reduce its financial commitment partly by increasing consumer prices for wheat products by about 26% between 1980 and 1984 -- still somewhat less than the rate of inflation -- and partly by purchasing wheat at lower prices. Except for 1988 and 1989, international wlheat prices have fallen during the 1980s so that Algeria has had to pay less for its imported wheat. Based on the existing exchange rate, the cost of this imported wheat in Dinar terms has fallen in all years except 1988 and 1989. The combination of these two factors is such that subsidy levels on imported wheat fell from about DA 400/ton in 1982 to negligible levels in 1986; indeed, Government showed a profit on imported soft wheat in 1986. With higher wheat prices in 1988 and 1989, subsidy levels again increased. 9.20 The subsidy for locally-produced wheat, on the other hand, is a function of the high producer and low consumer prices. In 1986, the subsidy on lccally-produced wheat was about DA 1,300/ton, and in 1989 about DA 2,500/ton. Grains purchased by the cooperatives (CCLS) on behalf of the national marketing bureau (OAIC) are sold to the mills, and the difference between the price paid to the producers and the base price to the mills, the "prix de base de rdtrocession" is paid directly to OAIC. A similar system exists for milk production, a small amount of which is purchased locally and mixed with lower-cost imported powder, for a production cost above the ex-factory price, Part of the difference is recovered from higher cost dairy products, and part from subsidies paid to the national milk marketing and processing office. Table 17 below summarizes the major subsidy payments in the agricultural sector to consumers and producers. 9.21 The CCLS generally purchase, from local producers, around 5-700.000 tons of wheat a year and an additional 150-350,000 tons of barley for a total of 650-1,000,000 tons of cereals. Based on the 1989 producer prices for cereals and pulses, which raised the price of hard wheat from DA 2700/ton to DA 4000/ton, producer subsidies can, therefore, expect to be some DA 1.5-2.0 bn and this assumes no supply response to the new, higher producer prices. With such a response, producer subsidies could exceed DA 2.0 bn (US$335 mn), or about 2.5% of expenditures on Government's current account (excluding expenditures on capital and financial accounts)'. This is not insignificant. As local production increases (Chapter IV), the quantity of wheat and other cereals to be purchased by Government will also increase. Assuming that farmers do not increase the quantity of wheat retained at farm-level, Government may end up purchasing 2.2 million tons of locally produced wheat and other cereals by the year 2000 unless the present policy is changed. Based on existing subsidy '/ In 1988 and 1989 current expenditures amounted to about DA 76 bn.  -73 - levels, this would require budgetary support of DA 5 bn, representing some 5X of projected Government recurrent expenditures of around DA 100 bn in the year 2000 (in 1989 prices). The issue is whether the benefits of the subsidy in terms of (a) supply response and therefore, foreigr. exchange saved, and (b) increased security are worthwhile in terms of the budgetary impact. This will be one of the issues addressed in the prices and incentives study. Table 17: FOOD AND AGRICULTURAL SUBSIDIESI (DA MILLION) 1982 1984 1986 1987 A. Soution des Prix Cereals 2760 1400 1340 2/ Imported flour 560 175 - - Vegetable Oil 530 400 340 Tntal 3850 1975 1680 B. Fonda de ComDensation (i) Consumer items - Cereals - - - 982 - Milk 285 209 116 300 - Sugar - 57 29 100 - Vegetable Oil - - - 265 - Olive Oil - Z 25 Subtotal 285 266 145 1672 (ii) Aariculture Inputs - Animal Feed - 497 225 490 - Fertilizers - 338 309 340 - Equipment and Machinery - 544 181 - - Pesticides - 31 - - Subtotal 285 1410 715 830 Total A + B 6135 3652 2540 2502 1/ The data in this table do not always correspond to lovols of rubaidios shown as received by the concerned parastatal. The data should, therefore, be takon as illustrativo of orders of magnitude. _/ Merged into Ponds de Compensation Source: Planning Ministry 9.22 The high prices of cereals (and other controlled crops), established each year by a Comit6 des Prix within the Ministry of Agriculture, are maintained by payment of subsidies from, in the past, the Soutien des Prix or, more recently, from the Fonds de Compensation. Allocations under Soutien des Prix were paid on cereals, imported flour and vegetable oil, with cereals accounting for about 50-80% of total subsidies. Resources for the Soutien des prix were made available by direct budgetary transfer. In 1987 the Soutien des Prix was merged with the Fonds de Compensation. This can be seen in Table 17. 9.23. The Fonds de Compensation was started in 1982 to finance. consumer subsidies on other products (milk, sugar, and olive oil) as well as, since 1983, subsidies on major agricultural inputs (fertilizers, animal feed, agricultural equipment and machinery and, for one year only, pesticides). The Fonds de Compensation is financed by a special tax called "taxe compensatoirel which is applied on a whole range of largely luxury goods. Overall, the amounts paid out from the two funds for agricultural related supports and subsidies have been around DA 2.5 bn (US$415 mr.). The amounts have decreased over time. As discussed in para 9.05, Government has stated its objective of phasing out subsidies on agricultural inputs by the end of 1989. This would reduce the  -74 - annual payments from these funds by about DA 800 mn (US$130 mn). However, the introduction of significant premiums for domestically produced wheat, which will have to be paid from the Fonds, will increase the annual payments from the Fonds. This increase will more than offset the reduction from phasing out subsidies on inputs, so that increased funding will be required for the Fonds de Compensation, unless, of course, the Government decides to reduce the level of consumer subsidies (para. 9.25). Increased funding would likely come from increasing the range of goods or the tax rate on the imported "luxury" goods. 9.24 These subsidies are the direct result of Government's policy to provide low cost food to urban areas, to "encourage" production of basic staples that otherwise have to be imported, and of the exchange rate. While it is important to ensure that basic foodstuffs be available to the population at affordable prices, it is equally important to ensure that the subsidies are serving this purpose. It has been found elsewhere, for example, that subsidized food programs frequently benefit those sections of the population that need subsidies least, the middle class urban consumer, and do not reach the people who do need support to ensure an adequate food supply. In addition, given the potential to reduce waste by increasing prices, particularly of bread products, reduced subsidies might also decrease consumption levels and reduce the level of dependence on imports. With wheat and wheat related imports running at about 3 million tons a year, at an annual cost of about US$450 mn, or 5% of total imports, avenues that have scope to reduce imports should be explored. This is all the more important given that food (wheat) imports could increase to 5.6 mn tons by the year 2000. At that time, they would represent 8% of total imports. 9.25 Moreover, as the dinar is devalued further, Government will either have to raise consumer prices or increase subsidies. The former will reduce demand and the burden on the budget but may be difficult politically to implement except in a phased manner. If consumer prices are not raised and the dinar were to devalue by 60-70% in the near future, for example, the subsidy on imported wheat and wheat related products could amount to another DA 1.0-1.5 bn, 2% of expenditures on Government's current account. This would have to be added to the DA 1.5-2.0 bn that will be required for subsidies on domestic production now that the price of hard wheat has been increased to DA 4000/ton. While Government may be able to afford to increase subsidies to this extent, the total impact of producer and consumer subsidies is considerable. Both from efficiency and budgetary points of view, consumer prices should be raised in a manner that takes full account of poverty and of political realities by better targeting funds available for consumer subsidies. The linkages between producer and consumer prices and subsidies and their impact on the foreign exchange and domestic budget need to be better understood. It is recommended that the efficiency aspects of consumer pricing and subsidies be studied in the context of the pricing and incentives study. Monies from the Fonds de Compensation have also been used to subsidize consumer goods other than those in the agricultural sector: domestic fuel, cement, soap and detergents. This also has important ramifications on the budget.  -75 - I. ImRact of Macroeconomic Policy on Agriculture 9.26 The macroeconomic setting, including relative prices, the exchange rate, interest rates, policies on industrialization and protection levels, and the framework for parastatal and private investments and operations all have a major impact on agriculture. Most of these policies have been reviewed in previous sections as has the issue of the linkage between high (subsidized) producer prices, low consumer prices, foreign exchange requirements for imports, and the budgetary impact of each of these policies. This section deals primarily with the effect of exchange rate changes on agricultural production and profitability. 9.27 Although Algeria's dinar has been gradually devalued over the past three years (1987-1989), it is still, by most estimates, overvalued. The impact of this recent devaluation and future devaluation on agriculture can be divided into the probable effect on producer prices and input prices, and thus on supply, on consumers' purchasing power, and thus on demand, and on the size of the subsidies required on imported food items. For output prices, most are determined freely by internal market forces with the exceptions of the cereals, pulses, and selected export fruit crops, which are controlled; the presence of quantitative restrictions (QRs) on imports plays a major role in determining "free-market" prices. For the cereals and pulLes, the output prices are currently set without reference to border prices, and are, as noted, much higher than border prices at the official exchange rate. Thus, in the absence of lifting existing QRs on imports, devaluation would have little direct effect on the nominal level of producer prices and would not serve to increase production incentives. 9.28 Since over 75Z of the wheat and about 30X of the barley products are imported, the major impact of devaluation on these crops is the resulting increase in purchase price to the country in dinar terms. Devaluation would serve to increase the subsidy requirements unless consumer prices were to be raised. For export crops, to the extent that the producer prices are set with reference to export parity prices (this will be followed up by the pricing and incentives study), devaluation would enable parastatal marketing agencies to increase the prices and thus provide a greater production incentive. 9.29 The impact of devaluation on input prices would depend on the extent to which the policy of "verite des prix" (para 9.05) encompasses the financial ex-factory cost of the products (fertilizers, tractors). At present, the policy uses an ex-factory price which is usually controlled and does not cover the cost of production. Since devaluation would mean an increased cost of intermediate goods in the production of inputs and equipment for the agricultural sector, this increased cost would be passed on to the farmers if "veritd des prix" was meant to cover production costs. This would reduce the profitability across all crops, and would serve to reduce supply at any given price. The extent of this impact on supply would depend on: the proportion of imported material in the manufacture of the inputs and equipment for agriculture and the effect of increasing intermediate prices on the ex-factory price; the price elasticity of demand by farmers for equipment and input purchases; and the technical responsiveness to changes in the levels of input and equipment use (production response to fertilizers, for example).  -76 - 9.30 Devaluation, through its effect on the price level, would serve to decrease real incomes and purchasing power, and thus would decrease demand for food. This would be felt in those products whose income demand elasticity is highest (pulses have an estimated demand elasticity of 1.9), and least for products such as milk (with an estimated demand elasticity of 0.8). A change in the price level would also decrease the incentives provided by producer prices that remained the same in nominal terms. 9.31 On balance -- in the absence of lifting QRs -- devaluation, whether gradual or sudden, would somewhat reduce profitability of agricultural production through the impact on input prices, and would thereby modestly reduce supply, except for export crops, where production might increase if producer prices are tied to border prices. Nominal producer prices for the major crops, cereals, would not be affected. Combined with a reduction in demand from decreased purchasing power, it is difficult to predict the direction of the movement in consumer prices. Subsidy requirements on imported wheat and barley would increase substantially unless consumer prices were increased accordingly. 9.32 Currently, the presence of QRs means that the price of fruits and vegetables, for example, is determined by demand and supply from the local market. If QRs were lifted, supplies would increase. Following a devaluation, the prices of imported commodities would be higher than they are today (if imports were allowed) by the amount of the devaluation plus any protection. Unless both the impact of devaluation and the protection rate are large, the import prices would remain below present "market" determined prices, especially for beef, so that the lifting of QRs is likely to further lower the price of commodities that are not controlled. As noted in para. 9.12, agriculture in Algeria is highly profitable and a modest decline in the terms of trade for agriculture is unlikely to affect farmer profitability and production significantly. X. CONCLUSION 10.01 This paper has considered the potential for growth of Algerian agriculture and concluded that growth rates could be as high as 6.6% per year until 1995, under the most optimistic scenario, and 4.7% p.a. thereafter. For this to happen, all the factors would have to come together, with constraints that have existed in the past being quickly removed or minimized. For what is probably a more realistic scenario, a growth rate up to 1995 of about 4.6% could be achieved and thereafter of 3.5%. Whichever of the two scenarios prevails, EAC/EAIs will have to respond quickly and surely to incentives, and all inputs will have to be readily available in a timely manner, and of good quality. For the longer term, Government would have to establish institutions capable of serving the research and extension needs of the farmers, and would have to develop its capacity to establish priorities for investments and see that a proper incentive framework were established for both the private and public sector to direct investments wisely. Finally, to ensure that this growth were achieved at the least cost and with minimal harm to the environuent, Government would have to carefully consider the options for development and would have to change its pricing and incentive structure to orient the development effort. Under less favorable circumstances (lower case scenario), the growth rates would be considerably lower, at 2.91 p.a. up to 1995 and 3.0X p.a. thereafter. While  -77 - this difference may not be large for a given year, the cumulative impact of different growth rates over time is substantial and will have a large impact on the import requirements, self-sufficency ratios, and export capability by the year 2000. 10.02 Which of the scenarios is realized is largely a function of Government's agricultural development strategy -- Government's commitment to resolve the issues discussed in this report. This will determine how well the EACs will be able to manage their autonomy and their resources. Government's commitment to improving institutions that service agriculture, to increasing the efficiency of the input delivery systems, and to providing adequate incentives will play a critical role in allowing the FACs and the individual farmers to continue to increase their productivity. In addition to further policy and institutional changes, Government has a role to play in carrying out economically viable investments that are well-planned and executed (paras. 8.36-8.37), and thereafter in providing sufficient budget an" manpower to maintain them. Without these necessary ingredients, Algerian dependence on imports for basic foodstuffs will grow, and agriculture exports will dwindle even further. With these elements, by 2000 Algeria may reduce its dependence on imports and increase its agricultural exports. 10.03 Many issues raised in this report need to be followed up, discussed, and acted upon in the near future to enable Algeria to realize its potential for agricultural growth. Most of the reforms and policy actions discussed here are independent of one another, that is, for example, reforms on pricing policy do not depend on Government actions to strengthen the research service. Thus, many of the actions and studies recommended in this report call be implemented simultaneously. To the extent that reforms should be sequenced, it is felt those reforms which contribute to the short-term, quantum jump in yields that can been expected as a result of the dissolution of the DAS should be given priority: availability of foreign exchange for the import of agricultural inputs and spare parts, liberalization of input supplies and equipment repair and measures related to the availability of credit. Work should also start on institutional reforms and studies. Finally, planning for future developments in irrigation and the South should be continued and strengthened; investment could, however, be delayed should there be implementation or budgetary constraints.  - 78 - LIST OF ANNEXES AND WORKING PAPERS1 Annex 1 Matrix of Agricultural Reforms in Algeria Annex 2 After Land Reform: Implementing the Break-Up of the State Farms Annex 3 Subsector Note: Livestock Working Paper 1 Estimated Agricultural Growth Rates and Projections of the Food-Gap to the Year 2010. Workin& Paper 2 Price Supports and Subsidies in Agriculture (Paper 1); Linkages between Import Costs, Producer and Consumer Prices - Cereals and Milk (Paper 2) 'j Available separately.  - 79 - STATISTICAL APPENDIX I. Production. Area and Yields Production, Area, Yields A.1 Gross Domestic Product, 1969-88 A.2 Cereals: Area, Production and Yields, 1964-89 A.3 Fodder Crops: Area, Production and Yields, 1964-89 A.4 Vegetable Crops: Area, Production and Yields, 1964-89 A.5 Pulses: Area, Production and Yields, 1964-89 A.6 Vineyards and Orchards: Area, Production and Yields, 1964-89 A.7 Livestock Production, 1967-89 A.8 Irrigated Areas by Crop, 1980 and 1986 A.9 Variability in Staple Food Production (selected countries), 1961-76 Indices of Production A.10 Cereals: Area, Production and Yields, 1964 - 100 A.ll Fodder Crops: Area, Production and Yields, 1964 - 100 A.12 Vegetable Crops: Area, Production and Yields, 1964 - 100 A.13 Pulses: Area, Production and Yields, 1964 - 100 A.14 Vineyards and Orchards: Area, Production and Yields, 1964 - 100 A.15 Livestock Production, 1967/68 - 100 Graphs of Growth Rates by Crop A.16 Cereals and Vegetable Crops: Production 1964-89 A.17 Pulses and Fodder Crops: Production 1964-89 A.18 Wine Grapes and Citrus Fruit: Production 1964-89 A.19 Red Meat: Production 1967-89 Summary Growth Rates: Production, Area and Yields A.20 Growth Rates: Production 1964-89, 1980-1989 A.21 Growth Rates: Area 1964-87, 1978-87 A.22 Growth Rates: Yields 1964-88, 1979-88 Value of Agricultural Production A.23 Value of Agricultural Production, 1964-88 A.24 Graph of Growth Rate: Value of Production, 1986 Prices  - 80 - II. Tae o ctors (Socialist/Private) A.25 Agricultural Area by Sector: Socialist and Private, 1980-86 A.26 Production by Sector, 1980-86 A.27 Yields by Sector, 1972-86 A.28 Growth Rates of Yields by Sector of Activity, 1973-86 A.29 Land Holding and Livestock Ownership by Sector (Socialist/Private), 1985 and 1986 A.30 Income Statements of the DAS, 1980-87 III. Comoarative Tables A.31 Comparison of Cereal Yields for Algeria, Tunisia and Morocco, 1973-88 A.32 Comparison of Yields of Principal Crops for SelecLed Cou-ntries, 1984-86 A.33 Availibility of Basic Foodstuffs in the Maghreb, 1982-86, 1976-80, 1966-70 IV. Imports/ExRorts A.34 Imported Agricultural Inputs and Equipment, 1977-88 A.35 Principal Imports of Foodstuffs, 1970-88 A.36 Principal Exports of Foodstuffs, 1970-88 V. Distribution of Inputs and Eguipment by Sector (Socialist/Private). 1980-87 A.37 Fertilizer Use by Agricultural Sector, 1980-88 A.38 Seed Use by Agricultural Sector, 1984-88 A.39 Pesticide Use by Agricultural Sector, 1980-88 A.40 Purchase of Agricultural Equipment by Agricultural Sector, 1978-88 VI. Production and InRut Prices A.41 Producer Prices, 1969-89 A.42 Indices of Producer Prices, 1974 - 100 A.43 Prices of Inputs and Agricultural Equipment, 1974 and 1980-88 A.44 Nominal Index of Input Prices, 1974 and 1980-88  ALGERIA GROSS DCWESTIC PRODUCT/PRODUIT INTERIEUJR BRUT 1969-1988 GROSS DOMESTIC PRODUCTYPRODUIT INTERIEUR BRUT 1969-1988 (milliorns of dinars, current priceslprix courents) twilLions of dinars, 1930 prices/prix 1980) GDP-AGRICULTURE GDP-TOTAL PERCENT OF TOTAL GDP-AGRICULTURE GOP-TOTAL PERCENT OF TOTAL PIB-AGRICOLE PIB-TOTAL POURCEN' DU TO7AL PIB-AGRICOLE PIB-TOTAL POURCENT Du TOTAL