Madagascar - Recent economic developments and prospects (Vol. 2 of 2) : Annexes

Report No. 2357-MAG Madagascar: Recent Economic Developments and Future Prospects (In Two Volumes) Volume 11: Annexes October 24, 1979 Country Programs Department Eastern Africa Regional pffice 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 duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Annual Average 1973 US$ 1.00 FMG 222.70 FMG 1.00 US$ 0.0045 1974 US$ 1.00 FMG 240.50 FMG 1.00 US$ 0.0042 1975 US$ 1.00 FMG 214.32 FMG 1.00 US$ 0.0047 1976 US$ 1.00 FMG 238.98 FMG 1.00 US$ 0.0042 1977 US$ 1.00 FMG 245.67 FMG 1.00 US$ 0.0041 1978 US$ 1.00 FMG 225.64 FMG 1.00 US$ 0.0044 Currency Unit = Malagasy Franc (FMG) FISCAL YEAR January 1 - December 31 WEIGHTS AND MEASURES Metric  FOR OFFICIAL USe ONLY MADAGASCAR RECENT ECONCKIC DEVELOPMENTS AND FUTURE PROSPECTS VOLUME 2 Note: The text of the Annexes comprising Volume 2 reflects the discussions with the Government of the draft report held In Antananarivo during June 11-15, 1979. The statistical information has, however, not been systematically updated. I Thi docmnt haa relotd dtIbubIon and may be med by cipients only In the performance 1 of thei elku duti. Its contents ay not othrwou be dIclos without World Snk authoriastion.   TABLE OF CONTENTS (Volume 2) Page No. ANNEX I - AGRICULTURE - SOME PROBLEMS AND POLICY ISSUES ............. 1 Introduction .................................................. 3 Commodity Developments and Issues ......................... 4 Strategy and Policy Options ........ ....................... 19 Appendix - Agriculture and Rural Development- Sector Memorandum . ............................ . 24 ANNEX II - INDUSTRIAL DEVELOPMENT IN MADAGASCAR ..................... 55 Statistical Tables ............................................... 65 ANNEX III - EMPLOYMENT ............................................. . 73 Current Trends .............................. -.0..... 74 Government Policies .. ............................ 81 Appendix I - Criteria for Sectoral Division of Employment . 90 Appendix II -Statistical Tables .......... .............. 94 ANNEX IV - PUBLIC FINANCE - MONETARY AND FINANCIAL STRUCTURE ........ 119 Summary and Recommendations ........................ 119 Monetary and Financial Situation ..o........ ........... 123 Activities of Financial Intermediaries .................... 130 Situation of the Major Financial Intermediaries ........... 133 Status of Public Finance .................................. 140 Monetary and Financial Plan for Madagascar ...............- 153 Appendix - Statistical Tables ........................ 162   ANNEX I AGRICULTURE SOME PROBLEMS AND POLICY ISSUES  -2- ANNEX I AGRICULTURE - SOME PROBLEMS AND POLICY ISSUES Table of Contents Page A. INTRODUCTION .............................................. 3 B. COMMODITY DEVELOPMENTS AND ISSUES ......................... 4 Rice ..................................................... . 4 Edible Oils .... ............................................ 8 Sugar ..................................................... 10 Cloves ................................. . .......... 13 Pepper ............ .................... 14 Vanilla .......................................... 15 Coffee .......................................... 17 C. STRATEGY AND POLICY OPTIONS ............................... 19 APPENDIX Agriculture and Rural Development-Sector Memorandum ...... 24  - 3 - ANNEX I Page 1 AGRICULTURE - SOME PROBLEMS AND POLICY ISSUES A. INTRODUCTION 1. This study was undertaken primarily with the intention of adding to existing Bank knowledge of the agricultural sector. It is accordingly selec- tive and concentrates on problems relating to some important agricultural commodities, particularly export crops, on the one hand, and on certain policy issues, on the other. Its findings and recommendations supplement those set out in a Sector Memorandum issued in June 1978; a shortened version of this memorandum is included in this Annex as Appendix. 2. Agriculture is in many ways the most important sector of the Malagasy economy. Crop farming, animal husbandry and allied activities together contri- bute about 40 percent of GDP. Its share in the GDP, however, does not fully reflect the prominence of the agricultural sector, which employs about 85 percent of the national labor force, meets over 75 percent of the household commodity consumption needs and accounts for, directly or indirectly, over four-fifths of export earnings. Unlike in a number of other developing countries, agricultural production is highly diversified; the product range is uncommonly wide and includes staple foods, industrial crops and a variety of export commodities. Paddy, however, is the leading crop and accounts for about 30 percent of agricultural production. Animal husbandry is a major rural occupation. Cotton, groundnuts, and sugarcane are the important indus- trial crops. Coffee, vanilla, cloves and pepper figure among the main exports, and together with sugar, made up about 70 percent of the total value of exports in 1977. Agricultural production techniques range from traditional cultivation to mechanized farming; but smallholder subsistence agriculture and traditional animal husbandry predominate in the rural areas. 3. Yields of food and other crops, however, are poor owing essentially to the prevalance of unimproved traditional cultivation methods that are characteristic of Malagasy agriculture. Lack of good crop production and acreage data make it difficult to assess productivity trends very closely. However, it appears that in recent years output per hectare of some important crops, like paddy, has tended to decline. Consequently, the growing demand for agricultural commodities has led to continued extension of cultivation to new lands; and cropped area has been rising at an average annual rate of some 3 percent. Agricultural production, however, has not risen significantly since 1975; and even since the beginning of the 1970s growth of output has averaged only about 1 percent per year, or less than half the population growth rate. The poor performance of the agricultural sector is seriously constraining the expansion and development of the economy. It has necessitated increasing reliance on food imports, particularly imports of rice and edible oils, which now preempt a significant portion of available foreign exchange that could otherwise have financed imports of scarce inputs and equipment. It has, besides, affected the availability of materials for processing industries and progressively eroded supplies for exports of such products as meat, meat preparations, quality rice and lima beans (pois du cap).  - 4 - ANNEX I Page 2 4. In 1978 Madagascar launched a three year Plan, which is to be the first of a series of medium-term investment programs intended to meet the essential consumption needs of the population and achieve rapid self-sustained economic growth by the end of the present century. Long-term planning also emphasizes state control over strategic activities, a more egalitarian income distribution, narrowing of regional disparities and decentralization of the economy through the development of local institutions (fokonolona) and coopera- tives. 5. The Plan envisages annual growth, in real terms, of GDP at 5.5 percent over the three-year period ending in 1980. Industry, which is to be the leading sector, is expected to expand at the rate of 10.7 percent, followed by mineral and energy sectors with an estimated growth rate of 8.8 percent. The Plan has also set a 4.3 percent annual growth target for the agricultural sector. For the attainment of production and other objectives, the Plan provides for a total gross fixed investment outlay of 184.3 billion FMG (in constant 1976 prices) during 1978-80. Even though the crucial role of agricul- ture in development is recognized, its share in the proposed investment program of 15.1 percent is smaller than the allocation for all other sectors except trade and distribution. 6. The likely impact of the Plan on employment is difficult to assess owing to the lack of relevant information about proposed programs and projects, particularly in the agricultural sector. According to the estimates set out in the Plan, however, employment likely to be generated in agriculture and the rest of the economy, about 65,000 jobs per year, will still leave about a third of the annual additions to the labor force unemployed. 7. The approach to agricultural development adumbrated in the Plan emphasizes rehabilitation of the irrigation system, modernization of farming, strengthening of extension services and the importance of producer prices as an incentive to farmers. It also envisages abolition of large private estates and the development of cooperative farming. However, family farms will continue to predominate, at least over the Plan period, and are expected to account for about 80 percent of value added in the agricultural sector during 1980. 8. This annex is divided into two parts. The first is devoted to the analysis of developments and issues relating to the main agricultural commodi- ties or manufactures based on them. The strategic and policy aspects of the problems of agricultural development in Madagascar are discussed in the second. B. COMMODITY DEVELOPMENTS AND ISSUES Rice 9. Rice is the main staple of the population and paddy the predominant crop in Madagascar. As with crop farming generally, traditional cultural techniques and practices prevail in paddy cultivation, which is characterized  - 5 - ANNEX I Page 3 by rather low yields. Agricultural statistics in Madagascar, including paddy and rice production estimates, leave much to be desired. It is thus difficult to ascertain production and availability trends accurately. It appears, however, that paddy yield has fallen off in recent years, and, consequently, production has been increasing rather slowly even though the area under cultivation has continued to expand in response to rapid population growth. According to official estimates harvested paddy acreage has increased since 1970 on an average by 3.4 percent annually. Output per hectare having de- clined, however, from 2.0 to 1.6 tons, paddy production recorded a growth rate of just about 1 percent, well below the 2.5 percent annual increase in popula- tion. 10. From 1.9 million tons in 1970 output of paddy increased only to 2 million tons in 1977, the estimated availability of domestically grown rice falling, over the same period, from 162 to 149 kg per head. As production failed to keep pace with population, Madagascar, which in normal crop years used to be a significant rice exporter, turned into a net importer of rice after 1971. And despite an increasing inflow of imports per capita consump- tion has tended to decline. Table 1: RICE: PRODUCTION, IMPORTS AND CONSUMPTION * Rice Net Apparent Production Apparent Consump- Production Imports Consumption per capita tion per capita -------'000 tons ----- - - -- - - (kg) - 1971 1,121 -12 1,109 162 160 1972 1,114 24 1,138 158 161 1973 1,071 44 1,115 148 154 1974 1,072 79 1,151 145 155 1975 1,097 72 1,169 145 154 1976 1,124 55 1,179 145 152 1977 ** 1,184 82 1,266 149 159 * Two year moving averages. ** Provisional. Source: Raw data - Direction Generale du Plan and INSRE 11. During most of the 1960s as well paddy cultivation had spread faster than populatiion growth. Comparable data reveal that between 1963 and 1970 the area under paddy increased by 3.55 percent a year. But over this seven year period output per hectare also went up at an annual rate of about  - 6 - ANINEX I Page 4 2 percent, whereas in recent years, particularly since 19-73, there has been a disquieting drop in paddy yields. The declining productivity of paddy farming can, in the main, be attributed to two factors: growing scarcity of goor ; croplands and the weakening of extension services. Since accessible and fertile valleys were already being intensively cropped, paddy cultivation w^s extended to inferior or less suitable lands, as is indicated by the fact L.tLI. the incremental yield of paddy during 1970-7.7 was much less than one ton per hectare. Several regional extension agencies provide services to paddy farmers in parts of Madagascar. Although formally under the Ministry of Rural Development they tend to function quite autonomously. Most were part of the GOPR (Groupement d' Operations des Productions Rizicoles), a large extension- project, now discontinued, which was financed by the FED. The extension organizations include the URER's (Unite regionale d'expansion rurale) of Itasy, Tananarive, Antsirabe, Ambositra, and Fianarantsoa, the ZER (Zone d'expansion rurale) of Ambilobe, and ODEMO, a development agency for the Middle West. An on-going UNDP rural development program in the south,is developing its own basic extension organization and services (Operation, Androy). Since about. 1972 the effectiveness of extension has suffered for the lack of adequate funds and technical personnel. As compared to an outlay of 30 billion FMG by the agencies constituting the GOPR during 1964-72 to finance intensive extension and improved farm inputs only about 7 billion would have been spent by them during 1974-79. Moreover, apart from smaller budgets, there has been a sharp reduction in the levels of foreign technical assistance received by them. 12. In its attempts to encourage paddy production the Government has relied essentially upon the price policy instrument. Following successive increases effected since 1972 the producer price of paddy went up from 15 to 35 FMG/kg; and as other crop prices were raised much less, relative prices moved substantially in favor of paddy. Little, if any, information is avail- able about prices actually paid by farmers for commodities and services. But judging by such indicators as retail price movements in urban centers, the purchasing power of paddy in terms of inputs and consumption goods too probably improved. The inducement offered through administered prices to raise paddy production and yields, however, does not appear to have been effective. This is not surprising; and the reasons for the inefficacy of the price mechanism are not far to seek. The adoption of improved farming techniques being constrained by the inadequacies of extension services or insufficient avail- ability of inputs - fertilizer imports have fallen by more than 50 percent since 1970 - the price incentive could, as it very probably did, only encourage farmers to bring comparatively marginal lands into cultivation or to dissuade them from switching to other crops. 13. Most rice production is consumed by farm families; surpluses are sold in good harvest years, with the bulk of marketed production coming from a few large irrigation schemes. Rice marketing and processing were in the hands of private traders with minimal Government intervention until the Revolution of 1972. Subsequently, rice marketing was nationalized. From 1973 to 1977, the Societe d'interet national des produits agricoles (SINPA), a parastatal  - 7 - ANNEX I Page 5 company, had the monopoly of marketing - collection, imports, processing and distribution of a number of agricultural products, particularly of rice. SINPA's marketing operations encountered serious problems, however, with shortages of rice in many areas, increasing requirements for rice imports and high costs of all marketing activities. In 1977, the Government assumed direct responsibility for rice marketing, while different parastatal agencies have been assigned responsibility for rice marketing in different regions, SINPA retaining the marketing operations in the largest of them, shipments of rice to deficit regions or storage are centrally directed by the Government. It is too early to judge the effectiveness of this system in resolving market- ing problems which have been worsening since 1972. With the slow growth of production and serious marketing problems, the marketed surplus, or more precisely procurement of paddy by the state rice marketing monopoly has been declining. As a proportion of production marketed surplus fell from 16-17 percent in the early 1970s to 12-13 percent during the last two years. And as compared to an average of some 175,000 tons (in terms of rice) during 1970/71 it dropped to about 155,000 tons in 1976/77. Local procurement as a conse- quence has had to be increasingly supplemented by imports in order to maintain the public rice marketing and distribution system. Net imports of rice increased from 25,000 tons in 1971 to an estimated 105,000 tons in 1977, and in the latter year accounted for nearly 40 percent of the total availability for public distribution. 14. The Government had been subsidizing imported rice so as to bring the price down into line with the lower cost of locally grown rice. In 1977, however, the retail price of rice was significantly reduced and subsidization was extended to local rice as well. At present the price to the consumer is set at 55 FMG/kg although the cost, at the retail level, of rice procured within the country is estimated to be about 75 FMG/kg while the c.i.f. cost of imports ranges from 90 to 95 FMG/kg. The import subsidy is financed by the Treasury, but the subsidy on domestic rice is paid out from the surpluses accumulated by the FNUP (Fond National Unique de Perequation), the centralized commodity stabilization fund, from export and other operations. The subsidiza- tion of imports alone is estimated to have cost about 3.6 billion FMG; and the overall burden of maintaining consumer prices at a level well below costs is likely to become heavier in the future. For one thing, import requirements may increase further. For another, international rice prices and also marketing and distribution costs are expected to go up. The maintenance of the existing arrangement for subsidizing rice consumption would thus become increasingly problematic. 15. The Plan, projecting an average 4.3 percent annual increase in output over the level attained in 1976, has set 2.3 million tons as the production target for 1980. About 80,000 ha of paddy land are to be developed over the Plan period, but some two thirds of the targetted increase in produc- tion is expected from improved yields. The annual 2.8 percent increase in output per hectare, implicit in these projections, is to be achieved through greater utilization of fertilizers and improved seed as well as through more vigorous extension efforts. As regards rehabilitation and development of paddy lands, emphasis will also be placed on smaller projects for reasons of cost as well as to encourage local participation in implementation.  - 8 - ANNEX I Page 6 16. An improvement in productivity of the order that is envisaged appears, at first sight, to be well within the realm of possibility, particu- larly as existing yields are lower than those that could be obtained even with traditional techniques. But in view of recent productivity trends and modest additions to output from extension of the cropped area, paddy production, until about 1980, can be expected to increase at about much the same rate as in the past few years. But if the efforts to improve extension are successful and a sufficient number of small short gestation irrigation projects are implemented, production may be expected to pick up further beyond 1980. Yet even if production expanded thereafter in step with population growth, annual import requirements by mid-1980s may reach 150,000 tons owing to the likely sharp increase in the rice deficit until 1980, a prospect which underscores the need for a longer run approach to the rice problem. In this context applied research as well as strengthening of links between research and extension are of crucial importance. Agricultural research has been centralized since 1972. However, CENRADERU (Center for Agricultural Research for Rural Development), the semi-autonomous body responsible for all agricultural research (including socio-economic analysis) has few formal links with extension services and suffers from a severe lack of qualified staff and funds. The Government has recently requested UNDP assistance for CENRADERU. The most critical area where assistance is required is in programming of applied research in rice production, notably field trials of new rice varieties. Edible Oils 17. Production of groundnuts, the main source of edible oils, has been stagnating for well over a decade. Except for the significant increase occurring during 1976/77, annual output of groundnuts has fluctuated around 40,000 tons. As groundnuts are grown on comparatively poor soils, with cultivation methods that have undergone little improvement, yields have been virtually static at the very low level of one ton per hectare. Production of groundnuts having remained more or less stable, growing demand since 1970 has necessitated increasing reliance on imports of fats and oils. Between 1970 and 1976 per capita apparent consumption of edible oils (including the oil equivalent of groundnuts consumed directly) went up from 2.3 kg to 3.9 kg and imports nearly trebled to 16 thousand tons.  - 9 - ANNEX I Page 7 Table 2: GROUNDNUTS AND EDIBLE OILS: PRODUCTION, IMPORTS AND CONSUMPTION * Groundnut Apparent Apparent Consupm- Production ** Import of Consumption tion per capita (oil equivalent) Fats & Oil ---- ---- (kg) ------------- ------('000 tons)-------- 1970 10.1 5.5 15.6 2.3 1971 9.4 11.5 20.9 3.0 1972 11.8 8.8 20.6 2.9 1973 9.0 8.1 17.1 2.4 1974 10.1 11.6 21.7 2.9 1975 11.1 14.7 25.8 3.4 1976 14.4 16.0 30.4 3.9 ** Net of exports which were discontinued after 1975. * Source: Raw data: Ministry of Rural Development and Agrarian Reform and INSRE. 18. The prolonged stagnation of groundnut production appears, at first sight, to be inexplicable, particularly as producer prices have been raised almost every year since 1970. From 24 FMG/kg in 1970, the producer price went up to 45 FMG/kg by 1976, and at present exceeds the corresponding border price equivalent by 10 FMG/kg. Currently fixed at 278 FMG/kg the ex-factory price of refined groundnut oil too is significantly above the estimated import parity price of 209 FMG/kg. It is possible that the producer price, given the very poor yield, did not always provide a sufficient incentive to induce farmers to bring more land under the crop. The farmers are reported to prefer cotton as an alternative to groundnuts, while the dislocation of marketing and transportation in recent years, might have further discouraged extension of cultivation for the market. Very probably it is for these reasons that the cropped area, barring the exceptional rise to 54,000 ha in 1976/77, continued to vary around a stable trend level of 40,000 ha. 19. With growing on-farm consumption the marketed surplus of groundnuts has fallen sharply in recent years. It dropped from 8.1 thousand tons (in terms of oil equivalent) in 1971 to only 4.5 thousand tons in 1976. As a result the availability of fats and oils for market distribution in urban areas increased only marginally over this period despite increasing inflow of imports; total availability for distribution rose by just about 5 percent to 20.5 thousand tons, while availability per head actually declined. 20. The Three-Year 1978-80 Plan has fixed the groundnut production target at 50,000 tons for 1980. No increase in yields is expected over the Plan period and the target is to be attained entirely by means of bringing an  - 10 - ANNEX I Page 8 additional 11,000 ha into cultivation. The projected expansion of groundnut area falls within the framework of the long-run development program for edible oils that is intended to help achieve self-sufficiency by 1990. The program envisages further expansion-of groundnut cultivation to 64,000 ha. as well as the development of copra oil and palm oil production to the level of about 11,000 tons. It is, however, doubtful that self-sufficiency will be achieved unless there is a marked improvement in groundnut yields. At the pre'sent low yield of one ton per hectare, output of edible oils (and oil equivalent of groundnuts consumed directly) could at best be expected to reach 28',000 tons by 1990, which would fall short of demand in that year by some 10,000 tons even if per capita consumption did not rise above the existing level. As such, the edible oils deficit, and hence imports, could be expected to decline only gradually over the next ten or twelve years, or to even increase if the objectives of the long run edible oils program were not fully attained. 21. Clearly, the major emphasis of attempts to raise oil-seed production should be on improving the productivity of groundnut cultivation, the more so as the scope for achieving higher yields is substantial. For instance, according to the Operation arachide, the support organization for groundnut cultivation, up to 2.5 tons per hectare can be obtained with better cultivation methods, varietal improvement and adequate extension'support to farmers. Sugar 22. The sugar industry in Madagascar consists of four refineries: two very old, small factories, producing some 25,000 tons' between them, and two comparatively large and modern plants. There has been no Lnvestment in new capacity for several years; and annual sugar production has been varying around an average level of 105,000 tons since 1970. With production more' or less stable and growing domestic demand for sugar, the volume of exports has been falling over time. Apparent internal consumption of sugar increased between 1970 and 1977 by 61'percent to 83,000 tons, the apparent consumption per head rising from 7.6 kg to 10.4 kg. Over the same period export fell by nearly-a half to an estimated 26,400 tons. Table 3: SUGAR: PRODUCTION, EXPORTS AND CONSUMPTION -Apparent Apparent Consump- Production Exports Consumption tion per capita --------------('000 tons)------------- ----- (kg) ------ 1970 101.6 50.0 51.6' 7.6 1971 93.3 33.4 59.9 8.7 1972 103.8 36.9 66.7 9.4 1973 88.2 37.3 61.9 8.6 1974 115.0 13.6 101.4 13.7 1975 114.5 39.9 74.6 9.9 1976 107.3 30.0 77.3 10.0 1977 109.4 26.4* 83.0 10'.4 Source: Raw data: Ministry of Rural Development and Agrarian Reform and INSRE'. * Provisional  - 11 - ~~~~ANNE LAI Page 9 23. The sugar pricing policy has been largely responsible for the stagnation of refining capacity as well as for the declining volume of exports. Until it was raised to 54 FMG/kg in 1977, the ex-factory price of sugar released for domestic market sales had remained pegged at 36 FMG/kg for about 25 years. While the retail price of sugar stayed unchanged at 70-73 FMG/kg during 1970-77, other consumer prices went up about 70 percent. The resulting steep fall in the relative price of sugar explains, to a large extent, the increase in per capita consumption at an average annual rate of more than 4 percent over a period during which real income per head declined signifi- cantly 1/. Besides encouraging domestic consumption, stable sugar prices inhibited investment in new capacity. The ex-factory price had initially been fixed at a level that was significantly higher than the price realized from export sales. But over time the profitability of the industry was progressively reduced as processing costs increased. And when world sugar prices reached exceptionally high levels during the recent boom, profits from export sales were largely mopped up by the levy of a 60 FMG/kg export tax. The industry was thus left with little incentive or resources for expanding capacity, and uncertainty about its future ownership until the nationalization of December 1976 discouraged investment further. 24. The Government intends to rehabilitate the two smaller sugar factories and estates and is seeking external financing for this purpose. The proposed project is expected to raise their production by 10,600 tons over the next seven years through improved sugarcane cultivation and replacement of old refinery equipment. With the implementation of the project sugar production is expected to increase to about 118,000 tons by 1985. 25. The Government increased the ex-factory sugar price from 56 to 75.5 FMG/kg in September 1978. The resulting increase in the consumer price of sugar, bringing it more in line with prices of other consumer goods, can be expected to restrain the growth of per capita consumption. And, as real income per head is not likely to go up significantly, at least until about the mid 1980s, aggregate consumption of sugar will, in all probability, rise more or less in step with population. The availability of sugar for export, however, would continue to decline even if consumption were to grow no faster. The exportable surplus may by 1985 be reduced to 18,000 tons and exports may practically cease by 1990. Table 4: SUGAR: PROJECTED OUTPUT, CONSUMPTION AND EXPORTS ('000 tons) Domestic Production Consumption Exports 1977 109.4 83.0 26.4 1980 110.6 87.1 23.5 1985 117.5 100.5 17.5 1990 117.5 116.5 1.0 Source: Mission estimates. 1/ The price elasticity of demand for sugar in Madagascar is estimated to be -0.62.  12 - ANNEX I Page 10 26. The prospect of sugar manufacturing eventually becoming an entirely domestic market oriented industry raises the question of desirability of investment in refining capacity and sugarcane plantations to build up exports again to the 50-60,000 tons level of a decade ago. No such program figures in the 1978-80 Plan which includes, besides the IFC rehabilitation project, a refinery-sugar estate complex (1,800 ha or say 18,000 tons annual production of sugar) to be completed at some future date in Morondava as well as two small open pan crystal sugar factories with a total capacity of 3,600 tons. 27. Even if a much larger eventual expansion of production were planned than is envisaged in the Plan, Madagascar would be in a position to secure markets abroad for the increased volume of sugar becoming available for export. As it is, the new International Sugar Agreement has alloted a basic tonnage quota of 70,000 tons to the country. The Agreement aims at dampening the fluctuations in the world sugar prices through the lowering of basic export quotas during periods of excess supply, and releases to the world market from stocks during periods of excess demand, like other small producers, Madagascar will be exempted from any restrictions on exports up to the basic quota of 70,000 tons. Besides, under the Lome Convention it has an annual quota of 10,000 tons for export to the EEC at a preferential "intervention" or guaran- teed minimum price. Exports, at least up to a fairly wide limit, are thus not likely to be constrained by the size of the market. As such, the desirability of production for the world market is very largely a matter to be decided with reference to the long-term viability of export sales. 28. Future profitability of additional sugar exports is difficult to establish with precision in the absence of information about production costs of a fair sized (60-70,000 ton capacity) new refinery-estate complex. Export production in the longer run, however, may at least break even financially. The cost of sugar cane (cultivated on irrigated land) can be expected to be about 8 US i/lb of white sugar.l/ Refinery investment cost, including contin- gency and escalation, of the 67,000 Dwanga Sugar project in Malawi is estimated at 437 US$ per annual ton 2/, or say total capital cost of 2.4 US 4/lb of white sugar assuming a twenty year asset life and a 10 percent interest rate. On the other hand, operating costs and administrative overheads together are expected to be about 2.0 US #/lb. Unit manufacturing costs of a similar sized plant in Madagascar should be more or less the same, so that the likely ex-factory cost of sugar would be about 12.4 US #/lb. As actual selling costs and freight amount to 2.4 US 4/lb 1/ sugar to be produced for export would have to fetch about 15.0 US #/lb cif Europe for costs to be fully recovered. 29. The world sugar price at present is much lower than-the likely cost of exports in the future. And even though the EEC guaranteed minimum price of 18.8 US 4/lb (cif Europe) is somewhat higher, it is applicable only to a comparatively limited volume of exports permitted under the quota. World 1/ Sucrerie Nossi-Be Cote East, S.A. IFC Investment Proposal October 1978 (IC/A/78-40). 2/ Swaziland Sugar Project. IFC Investment Proposal August 1977 (IC/A/77-29).  - 13 - ANNEX I Page 11 sugar prices, however, are forecast to rise to 16.5 US i/lb (in 1978 dollars) by 1985 and to remain at about the same level until 1990. The corresponding cif Europe price of 18.2 US #/lb for white sugar would be even higher. Thus, even though Madagascar is not among the low cost sugar producing countries, there is a distinct possibility for sugar to be a worthwhile export. The matter indeed merits to be explored in depth, particularly so as the profit- ability of exports will be significantly higher if foreign exchange and primary inputs are shadow priced to reflect their scarcity value to the economy. Cloves 30. Madagascar and Tanzania are the principal exporters of cloves, having roughly equal shares in the world market for this spice. Most of their exports, averaging in all about 14-15,000 tons annually, are destined for Indonesia, the largest single producer and consumer of cloves in the world. Exports to other countries (excluding Singapore and Hong-Kong which are essentially entrepot centers) have, on the other hand, amounted to about 3,000 tons per year. 31. The market for cloves has been expanding slowly. Imports into countries other than Indonesia did not, with the notable exception of France, register any significant increase during the 1970s. In 1975 Madagascar entered into an agreement with Tanzania to set export prices and to maintain them at a stable real level through the regulation of stocks held by the two countries. 32. The outlook for cloves has become somewhat uncertain as future market prospects are likely to be affected by Indonesia's efforts to achieve self-sufficiency. The demand for cloves in Indonesia, where they are used almost entirely for the manufacture of kretek cigaretts, is expected to exceed 33,000 tons a year by 1980. Local production has been fluctuating between 15 and 25,000 tons, but additional area planted with the crop would soon begin yielding. The eventual increase in production may not completely eliminate imports, partly because cloves of foreign origin will be needed for blending with the local variety; it will, nevertheless result in excess supply in the world market. 33. International prices of cloves, which have been rising since 1972, increased sharply in 1976. Average cif London prices of cloves from Tanzania and Madagascar went up by 28 and 34 percent respectively during the year, reflecting, among other things, steps taken by the two countries to control the flow of supplies into the world market. Prices since 1976 have continued to firm up further. 34. Production of cloves in Madagascar is concentrated on the east coast, north of Tamatave. The area under it has been increasing regularly since 1970, largely because the price to the grower (currently 320 FMG/kg) was sufficiently remunerative. As a result average annual production went up from 7,500 tons in 1970-73 to about 12,000 tons during 1974-77. Exports, however, did not register a similar rising trend; barring the exceptional surge to 22,500 tons in 1975, exports varied between 5,000 and 7,000 tons.  - 14 - ANNEX I Page 12 35. The 1978-80 Plan has set a production target of 16,800 tons, which implies a 7 percent annual growth rate over 1976, the base year for projections. The production target is to be attained essentially through measures to improve yields per hectare as the gestation period of cloves extends from 5 to 6 years. However, a program to increase the area under cultivation is being implemented, which may raise the annual growth of production to over 10 percent as new plantings begin to yield in the period beyond 1980. The problem, so far as exports of cloves are concerned, is thus not one of avail- ability but one of securing markets for expanding production. The production target and also the development program for cloves are, strangely enough, in conflict both with action to directly influence prices and with the uncertain market outlook resulting from Indonesia's attempts to eliminate dependence on imports. 36. Regulation of export sales, in concert with Tanzania, may be the appropriate policy in so far as the market is not expanding significantly. But steps now need to be taken to improve longer run prospects by a search for new markets and diversification into the production of quality essential oils and oleoresins. Pepper 37. Madagascar is a minor producer and exporter of pepper. Exports normally account for about 5 percent of the world market in pepper whereas India, Malaysia, Indonesia and Brazil are the leading suppliers. Most of the pepper from Madagascar is exported to the EEC, mainly to France, and to a much smaller extent, to the Federal Republic of Germany. 38. Although world imports declined significantly in 1974 and 1975, the outlook for the coming years is quite promising. Market prospects are improv- ing owing to the increasing use of pepper for extraction of oleoresins and essential oils required by the food processing industries. It is also being more widely used as a condiment. While demand in North America and Some West European countries is forecast to expand by 1 to 3 percent annually, the market for pepper in France and the Federal Republic of Germany is expected to grow at the rate of about 5 percent. Moreover, consumption in East European countries and Japan, which are significant consumers and together account for about a fifth of world imports, is projected to increase at annual rates of 5 to 10 percent. The consumption of green pepper (packed in water, brine or vinegar) though still limited, is rising. Madagascar is the leading supplier of green pepper, and France by far the biggest importer. 39. World pepper prices tend to be unstable due to fluctuations in supply from producing countries and, to some extent, as a result of speculative trading. Prices since 1972 have, however, been rising steadily. Pepper consumption is not very responsive to price variations; and fairly large ups and downs in prices appear to have a rather limited impact on world demand.  - 15 - ANNEX I Page 13 40. Production trends in recent years are reflected in the volume of exports from Madagascar which has fluctuated around 3,000-4,000 tons since 1972. The 1978-80 Plan aims at steeping up the production of pepper through, among other things, better extension and more intensive application of compost or fertilizers. Another Plan objective, with longer term implications for production, is to promote research into high yielding, disease resistant pepper varieties which are well suited to local conditions. However, along with measures to improve yields the price to the producer may need to be raised to a more remunerative level in order to attain faster growth of pepper output. Markets abroad for a growing volume of production should not be difficult to find for a number of reasons. First, demand is expected to rise comparatively fast in France and the Federal Republic of Germany, which are the main markets for Madagascar. Second, it should be possible to secure, through suitable pricing and sales initiative, a better share of the market in other countries. For, even if world demand is not price elastic, the market for a small producer, like Madagascar, can be quite responsive, although only up to a point, to prices quoted by it. Finally, the spread between costs and export prices is wide enough to permit sales, if necessary, at more attractive promotional prices in new markets. Vanilla 41. Madagascar is by far the world's largest producer of natural vanilla. The highly flavored Bourbon variety, grown like cloves in the northeast, accounts for close to 80 percent of vanilla entering the world market. The USA is the leading importer of natural vanilla followed, in the order of importance, by France and the Federal Republic of Germany. The three countries normally absorb the bulk of exports from Madagascar. 42. The market for natural vanilla was seriously hit by the development of synthetic substitutes, particularly the low cost vanilla derived from waste sulphite liquor and coal tar extracts. The annual production of substi- tutes, in natural vanilla equivalent, now exceeds 40,000 tons. Faced with declining demand for natural vanilla and seeking to maintain prices, Madagascar, in concert with major importing firms in the USA and Europe, began to regulate supplies into the export market in 1964. Exports over the last ten years or so have, as a result, remained fairly stable and averaged around 1,100 to 1,200 tons per year. 43. The demand for natural vanilla has been picking up in recent years and prices as a consequence have risen substantially. From US$5.65/lb in 1972 the average spot New York price of vanilla from Madagascar went up to US$9.25/lb in 1976 1/. The revival of demand for the natural flavoring is attributable largely to food legislation in important consuming countries and the promotional effort undertaken by vanilla producers. Sales of natural vanilla in the USA have been significantly helped by legislation requiring labels to indicate the type of flavoring used in ice cream products. In 1/ Since 1976 prices have risen further.  - 16- ANNEX I Page 14 France legislation introduced in 1967 curbed the use of synthetic vanilla in certain types of food products while labelling regulations were made more strict. Univanille, an organization of vanilla producers in Madagascar, Reunion and Comoros, has sponsored extensive promotional campaigns in both countries. 44. The market outlook for natural vanilla is promising. Over the five years 1971-75 world imports increased by a little over one percent annually but consumption in the future is likely to increase at a faster pace provided availability in the world market increases in step with demand. Besides, it is believed that adoption of food legislation, similar to that in force in the USA and France, in all EEC countries could raise world consumption by 600-800 tons, or say about 50 percent, over the next 8 to 10 years. 45. The production of vanilla has been falling off in Madagascar since 1975 leading to a continuing decline in stocks. The 1978-80 Plan aims at raising annual production again to the level of 6,800 tons (green vanilla) by 1980. This production target is to be attained through a number of measures including, among others, increase in the producer price, plant protection and disease control. The Plan, however, does not envisage any significant increase in the volume of exports. And the production program, accordingly, is intended only to replenish stocks so that exports can be maintained at the past average level of about 1,100 tons a year. No long term vanilla policy is set out in the Plan; but in view of the improved market outlook and prospect of a continu- ing upward shift in the demand for natural vanilla in the coming years, production possibilities as well as the export marketing strategy deserve to be considered in the longer term context. 46. New areas brought under cultivation begin to yield only after a period of about three years so that vanilla production can be increased only marginally in the short period, essentially through improved plant care. But there appears to be no physical factor to limit the growth of output in the period beyond 1980. The area under the crop is as yet small (estimated to have been about 25,000 ha. in 1977), and provided the price to the grower is attractive enough, production can be increased to keep pace with the expected growth of world demand and normal stock requirements of double the annual flow of exports. 47. So far releases for export sales have been maintained at levels that did not necessitate any marked permanent increase in vanilla production. Such a strategy was no doubt justified in a period when the market was shrinking as, apart from shoring up prices, it probably also raised total export earnings from vanilla over what they would otherwise have been. However, a more expansive marketing strategy would perhaps be more appropriate in the present circumstances. For one thing, as demand is rising and expected to expand further, static or inadequately increasing supplies for export are more likely to raise exporters' (and importers') profits than the value of vanilla exports. For another, a further rise in prices owing to insufficient increase in market availability entails, in the longer run, the risk of a renewed loss of markets to synthetic substitutes. Thus along with promotional efforts to widen the  - 17 - ANNEX I Page 15 market for natural vanilla, the export policy itself needs to be carefully reviewed, particularly with regard to the possibility of raising export earnings through a growing volume of exports and lower export prices. In any case, the export cost-price differential of FMG 2,000/kg appears to be large enough to leave a substantial profit margin even after allowing for reductions in unit prices resulting from the expansion of exports and such increases in the producer price as may be necessary to achieve the required increase in production. Coffee 48. Coffee is Madagascar's foremost export commodity accounting for about 25 percent of the country's total export earnings in normal years. Its share in total exports, however, rose an estimated 50 percent in 1977 as world coffee prices soared to unprecedented levels. 49. Coffee production in Madagascar rose at an annual rate of 5.2 percent between 1970 and 1977 owing largely to the efforts of Operation Cafe, the extension and development agency for coffee. Somewhat over half of the increase in output of 33,000 tons over this period occurred as a result of expansion of the area under coffee. Exports too increased fairly steadily although at a slightly lower average of 4.6 percent per year. Per capita domestic consumption of coffee, however, did not exhibit any definite trend; and the local market offtake, notwithstanding year-to-year fluctuations, increased at roughly the same pace as population. Table 5: COFFEE, PRODUCTION, EXPORTS AND CONSUMPTION Apparent Apparent Consumption per Production Exports Consumption capita ('000 tons ) ------ ---- (kg) ----- 1970 66.6 52.0 14.6 2.2 1971 57.7 51.9 5.8 0.8 1972 68.6 56.2 12.4 1.8 1973 73.9 65.4 8.5 1.2 1974 81.0 65.4 15.6 2.1 1975 83.6 67.4 16.2 2.1 1976 78.9 73.0 5.9 0.8 1977 89.3 64.0 1/ 25.3. 3.3 1/ Provisional: According to the most recent estimates coffee production in 1977 was only 68,000 tons due to poor climate conditions, and only 50,000 tons were exported during the year. Source: Ministry of Rural Development and Agrarian Reform and INSRE.  - 18 - ANNEX I Page 16 50. As yet the domestic market for coffee is relatively small absorbing normally less than a fifth of total production. There is, moreover, little likelihood of a marked increase in per capita demand, and so of a rapid growth of domestic coffee consumption, in the coming years. For one thing, following recent increases, the retail price of coffee has risen by about 42 percent. For another, a significant improvement in the per capita real income is not expected in the near future. Export prospects, accordingly, will depend, in the main, on world market conditions in the future, on the one hand, and production trends, on the other. 51. The recent coffee boom, triggered by the Brazilian frost of 1975 and subsequent running down of world coffee stocks, peaked out in 1977. But excessive plantings in response to the exceptionally high prices during the boom period would be bringing about a steep increase in world production in the 1980s. World coffee prices, therefore, are forecast to continue falling over the next four or five years and to recover only gradually thereafter. The International Coffee Agreement (1976-82), in which Madagascar participates as a member of the Organisation africaine et malgache du cafe, provides for the introduction of export quotas to limit market availability in line with the growth of world import demand in the event of prices falling below speci- fied levels. Quotas will probably be introduced in the near future so that annual growth of coffee exports from Madagascar may be held down to about 2.6 percent, the projected rate of expansion of world import demand during the early 1980s. Although quota restrictions, if put into effect, are unlikely to be maintained over an extended period of time, the growth of world demand itself is forecast to slow down to just about 0.6 percent a year beyond 1985. Thus, at first sight future world market trends appear to preclude the possi- bility of improving, or even maintaining, the pace of growth of Madagascar's coffee exports. Madagascar, however, is a minor coffee producer and claimed just about 2 percent of the world coffee market in 1976. Moreover, in the case of the arabica variety of coffee, for which soils in growing areas are better suited, the market share is even smaller. Madagascar should accordingly be able to plan for at least a 5 percent annual increase in the volume of exports over the long run without running the risk of provoking retaliatory price reductions by other exporters. 52. World coffee prices in real terms are projected to remain well below the current level throughout the 1980s. 1/ However, despite low yield per hectare of small holder cultivation -- 370 kg as compared to 600-700 kg in Kenya and Tanzania -- the cost of production in Madagascar should continue to compare favorably with the likely realization from export sales. According to estimates furnished by the Ministry of Economy and Commerce in June 1978, the fob cost of coffee (at the present producer price of 180 FMG/kg) was only 255 FMG/kg as compared to the fob price of 599 FMG/kg. If, as is likely, Madagascar's export prices declined in the future in step with projected world prices, the cost-price margin, barring an exceptional year or two, would still 1/ As a result of the reported recent severe frost in Brazil, price prospects for the early 1980s. may be much better than had earlier been expected.  - 19 - ANNEX I Page 17 remain sufficiently wide to permit a significantly higher price to be paid to the grower despite such promotional price reductions as might be offered to foreign importers. 53. The current Plan has set the coffee production target for 1980 at 104,000 tons, which implies steady annual growth of 4 percent over the average 1976/77 level of 84,000 tons. The Plan production objective is intended to provide a surplus of 80,000 tons for export after meeting the domestic consump- tion requirements. The increase in coffee production envisaged in the Plan, to be attained mainly through better and more widely diffused extension services, appears to be feasible. Strengthening of extension work, which has suffered from shortages of personnel and financial resources, may help maintain the annual 2.2 percent increase in yields while new plantations coming into production (the gestation period of the coffee shrub is four to five years) would raise output further. The domestic consumption target for 1980 of 24,000 tons, however, is rather unrealistic involving as it does a more than 50 percent increase over the 1976/77 average of 15,600 tons. Beyond the Plan period, particularly during the latter half of the 1980s, the growth of coffee production can be stepped up further to permit a faster expansion of exports and to meet increasing domestic consumption requirements. There is no scarcity of land suitable for coffee cultivation; nor is output growth likely to be constrained by other physical factors. The price to the producer may however have to be raised significantly to provide sufficient incentiVe to growers to expand cultivation. III. STRATEGY AND POLICY OPTIONS 1/ 54. A dynamic agricultural sector is one of the critical pre-conditions for more rapid and sustained growth of the Malagasy economy. Agricultural development accordingly has received high priority from the Government and the planners. Nevertheless, steps taken in the past few years to stimulate agricultural production have met with little success. Area or product oriented extension schemes, increase in producer prices of important crops and even the mounting of special programs to raise paddy production did not lift agricul- ture from the state of near stagnation. The unsatisfactory performance of the agriculture sector reflects partly the complex and deep seated problems confronting it. But the absence of an effective strategy has been an equally, if not more, important obstacle to agricultural development. It is in order, therefore, to consider the basic strategic and policy options in the context of Malagasy agriculture. 55. As yet, the soil resources have been only partially exploited, and just about a fifth of the cultivable land is at present being cropped. Even so, simply bringing more land into cultivation cannot be depended upon to provide for the growing requirements of basic foods and other agricultural commodities. While the area under food crops, particularly paddy and tubers, has been increasing in response to the rapid growth of population, crop yield by and large have remained static or even declined. Manifestly, the most 1/ See also Appendix in Annex I.  - 20 - ANNEX I Page 18 urgent need is to improve productivity; thus even as more accessible fertile lands are developed, the strategy should be to emphasise the diffusion of high yielding seed-fertilizer technology backed by adequate extension support, and, in the longer run perspective, on intensive cultivation and extension of irrigation. A basic extension program for key crops and estab- lishment of working links between research and extension thus merit a high priority since Madagascar as yet has no basic agricultural extension service, and current links between research and extension are weak. Extension services are principally provided by a large number of autonomous agencies and paras- tatals, organized on an ad hoc basis in the past. Also, in the context of action to improve productivity irrigation acquires particular importance as it offers even with traditional farming techniques, considerable scope for increasing both crop yields and gainful employment opportunities. Owing to high manpower input requirements, increased cropping intensity and better yields irrigated cultivation greatly extends the productive use of labor in crop farming. It has been found that under certain conditions, the introduc- tion of irrigation may, with traditional cultivation, double crop yields and employment per hectare. 56. Because of its sizeable employment potential alone, irrigation should constitute a key element of the long-term agricultural development strategy. For one thing, the agricultural sector has been absorbing the major portion of the annual of some 100,000 workers addition to the national labor force at rather low and diminishing output and incomes per head. For another, it appears inevitable that employment opportunities in the future would have to be created mainly within the agricultural sector. With the comparatively small industrial base and fast rising population, the growth of manufacturing, even if rapid, is unlikely to raise substantially the proportion of the labor force it employs except in the very long run. Planning Ministry projections reveal, for instance, that the expansion of industry, mining and energy sectors at an annual rate of about 9 percent would generate only 5,000 jobs per year. By the end of the century their combined share in total employment would rise to a little over 3 percent as compared to 1.7 percent at present. 57. While there can be little doubt about the desirability of further exploitation of water resources -- according to official estimates only about 184,000 ha are being cultivated under some type of irrigation as compared to about 900,000 ha of potentially irrigable land -- and the coverage of the entire cropped area by productivity programs with adequate extension support as long term objectives, the availability of resources and skilled personnel will impose definite constraints on the pace at which these objectives can be attained. As such, action along these lines will have to be, at least in the initial stages, carefully designed and phased. Priorities may thus have to be established between crops and types of farmers with regard to the pro- vision of irrigation improved inputs and extension services. In the present situation it would be preferable to direct resources and efforts primarily towards paddy and groundnuts. Paddy does, and with better yields groundnuts should, provide an opportunity for efficient import substitution. But, inadequate production and diminishing marketed surplus of these crops has made it difficult to meet the normal requirements of the urban population and short-falls in procurement have necessitated sizeable imports. For similar  - 21 - ANNEX I Page 19 reasons, that is to say, concentration of efforts in areas of greater compara- tive advantage, priority might also be given to better lands and farms, already producing a surplus for the market, instead of attempting to reach both market and subsistence farmers. The choice in this respect, however, is less clearcut as selectivity with regard to farmers will tend to make the income distribution in the countryside more unequal than selectivity with regard to crops. Ultimately such a choice would depend on the importance the Government income distribution and growth objectives, or the possibility of taxing the eventual addition to the incomes of richer farmers. In the matter of irrigation, priority will need to be accorded to the rehabilitation of the existing network whose capacity has been seriously impaired by the neglect of maintenance. Also as a step towards the rational development of irrigation facilities it would, be imperative to carry out an in-depth review of cost and technical aspects of feasible alternative techniques so as to avoid the installation of facilities involving excessive costs. 58. Efforts to raise crop yields and the marketed surplus will, however, be successful only if prices received by the farmers ensure an adequate return over production costs incurred by them. Producer prices of crops in Madagascar are regulated by the Government which provides it with an important policy instrument to influence agricultural production. The Government is aware of the importance of price incentives; and the producer price policy objective is to ensure a fair income to farmers 1/. It is also recognized that once fixed at a certain level, the producer price cannot be reduced below it. The application of these broad guidelines has, however, varied greatly from product to product. Sharper increases in the producer prices of paddy since 1972 have raised it substantially in relation to the producer prices of other crops; at the other extreme the sugarcane producer price remained pegged at the same level for years before it was revised in 1977. The producer price of paddy is significantly lower than the corresponding import price equivalent, while the producer price of groundnuts exceeds the import parity. 59. In view of the manner in which prices are set, pricing of agricul- tural commodities becomes an important policy issue w4ithin the framework of an overall sector strategy. The pricing policy ideally should be based on world market prices. Producer prices which approximate the corresponding border price equivalent provide, other things being equal, adequate incentives for replacing imports, or expanding export production, to the extent advantageous to the economy. The international parity pricing solution, however, raises some operational and other issues in Madagascar. The producer price of paddy is still well below the import price equivalent. As such, parity pricing, unless introduced by gradual steps, could well create serious problems for the Government, which has kept paddy and rice prices low in order to limit the increase in the cost of living, particularly of the urban poor. At the present, paddy producer price of 35 FMG/kg the cost at the retail level of local rice works one to about 75 FMG/kg. The uniform consumer price of rice, fixed at 55 FMG/kg, thus involves subsidization of local rice to the extent 1/ Premier Plan, 1978-80, Direction Generale du Plan, December 1977 p. 72.  - 22 - ANNEX I Page 20 of about 20 FMG/kg. The effective subsidy in the case of imported rice is much higher - about 40 FMGIkg at the current cif price of rice which ranges from 90 to 95 FMG/kg. Adjusting the producer price of paddy to the import parity level of 48-50 FMG/kg would thus result in a 70 percent increase in the uniform retail price of rice. In fact, the retail price may have to be raised much more since world rice prices (in constant dollars) are projected to rise by about 38 percent during the 1980s to reach what is considered the "normal" level. In the case of groundnuts, the producer price exceeds the import parity. There is, however, considerable scope for raising yields, and parity pricing would become feasible as comparative advantage is established through efforts to raise productivity. The situation with regard to sugarcane too is similar; despite having remained relatively stable, the producer price exceeds the corresponding international price. But, as world sugar prices rise to more normal levels, more optimal sized sugar refineries are establlshed and sugarcane yields improved the producer price could become internationally competitive. 60. As regards some important agricultural exports - coffee, vanilla, cloves, pepper - unit price realization normally exceeds production costs because of Madagascar's significant comparative advantage in these commodities. Even so producer prices may have to be kept below the export parity level. Because of uncertain prospects for cloves, securing markets for a growing volume of exports is likely to be problematic. As for coffee, vanilla and pepper, better price incentives to producers will be needed to increase export production. However, raising the producer price right up to the export parity level might encourage production beyond the point where the demand for exports from Madagascar turns inelastic. Among other agricultural exports, meat pricing raises issues similar to those encountered in the case of rice. In order to restrain the rise in the cost of living, beef prices have been fixed at a level much lower than the export price. 61. Clearly, flexibility will need to be exercised in shifting towards a more rational system of agricultural prices. In the first place, parity pricing will have to be introduced gradually, particularly when the disparity between producer and parity prices is rather large, or a sustained improvement in crop yields is necessary to bring the producer price into line with the border price. Second, care should be taken that in the process of adjusting to parity prices, input prices, particularly the price of fertilizer, are not raised to such an extent as to discourage optimal utilization of improved farm inputs. The present urea-paddy price ratio of about 3 : 1 is, for instance, significantly higher than the corresponding international price ratio. Third, wherever possible, initial increases in producer prices should be large enough to enable farmers to meet, in addition to operating and maintenance costs, at least some portion of the investment cost of projects from which they benefit. Finally, deviations from parity pricing may be unavoidable in the case of certain agricultural exports. 62. An important aspect of pricing policy relates to the subsidizing of rice, a commodity which figures prominently in the consumption of low income urban households. The subsidy imposes an additional burden on scarce budgetary resources and absorbs a significant portion of the reserves of the FNUP.  - 23 - ANNEX I Page 21 The cost of subsidizing rice imports (initially estimated at 75,000 tons by the Ministry of Economy and Commerce) as well as locally procured rice is likely to have amounted to about 7.0 billion FMG in 1978. At present all segments of the population benefit from the rice subsidy. In keeping with the Government's social objectives the subsidy ought to be limited to purchases by low income households. The well-to-do might also be made to bear the cost of subsidizing the poor in one of the two ways. Since rice procurement, imports and marketing are handled by state trading agencies, the subsidy might be financed by setting the retail price for high income consumers at a level that sufficiently exceeds the overall average cost of rice. The main drawback of this type of dual pricing arrangement is that it will need to be backed up by a rice ration program for the target group consisting of the poorer strata of the population. The costs of administering such a program will be considerable; there is also the risk of leakage of subsidized rice to the high price market. Alternatively, the cost of the subsidy could be met through additional taxation of personal incomes and luxury goods. This method has the great advantage of avoiding the need for rationing. However, given the likely scope for increasing tax revenues, it is very doubtful that necessary resources could be raised to finance the entire cost of the existing subsidy. On the whole, reliance on additional taxation to the extent feasible seems to be the most practicable method of subsidization through income transfers. For this reason, as also the exceptional character of the substantial profits from exports, which subsidize locally procured rice, the subsidy may have to be reduced sooner or later.  - 24 - ANNEX 1 APPENDIX Page 1 MADAGASCAR AGRICULTURE AND RURAL DEVELOPMENT - SECTOR MEMORANDUM 1/ A. INTRODUCTION 1. This memorandum is intended principally to provide general back- ground and policy recommendations for discussions between the Bank/IDA and the Malagasy Government on the agricultural and rural development sector, and to assist in formulating a lending program for agriculture and a plan for related project preparation activities. It is based on information obtained in the course of operational missions by Project Department Staff, most recently, appraisal of the Mangoky Agricultural Development Project in April 1978. It also draws on a variety of reports on Madagascar, notably the most recent IBRD economic report (December 20, 1976). Preparation of this memorandum was prompted by three factors: (a) New lending for agriculture has been virtually at a standstill for more than four years, and both Bank staff and Government officials have expressed dissatisfaction about past lending operations, several of which have encountered serious problems in implementation; we hope the time is ripe to engage in an active lending pro- gram for agriculture, as part of a coherent program which meets the objectives of both the Government of Madagascar and the Bank; (b) Dialogue between Bank missions and Government officials has been frustrated in the past year by conflicting views expressed by Ministry of Rural Development and Ministry of Finance and Plan officials on priorities among projects and general sector objectives. Discussion of objectives and specific review of potential proj- ects may help to advance the working level dialogue in the future; and (c) Considerable information is available on the agricultural.sector but in a rather fragmented form; provision of a general synopsis of such information should provide useful background for future discussions and operations. 2. Bank knowledge of the agricultural sector has been gleaned princi- pally from project operational activities and from analysis of agriculture undertaken as part of general economic work. It is thus somewhat biased towards the sub-sectors and geographic regions where we have been most active. No formal sector work has been undertaken in the past eight years, although an agricultural project identification mission in 1973, which was partially aborted, did gather considerable general sector information. 3. Changes in political orientation, general strategy, institutions, and personnel since 1972 in Madagascar have had an important impact on recent discussions and operations in the agricultural sector (and in other sectors as well). The change in Government in May 1972 was the first of a series of major upheavals which have effectively revolutionized Madagascar's poli- tical orientation and institutions. During this period of turmoil, which 1/ This memorandum was prepared by staff of the Central Agricultural Division, East Africa Projects Department in June 1978 and has been somewhat shortened for inclusion in the Economic Report.  - 25 - ANNEX 1 APPENDIX Page 2 was marked until mid-1977, and which continues to a much lesser extent at present, there was a general resistance to external influence and advice, while the atmosphere of political uncertainty discouraged civil servants from taking strong action, prompted frequent staff changes in some institutions, and generally resulted in more cumbersome and slower bureaucratic procedures than had previously been the case. We have been reticent to move aggressively to propose projects or to assist in their preparation while the Government told us, in effect, to wait until they were better prepared to discuss their own priorities. Another important change has been the rapid replacement of Madagascar's expatriate technical assistants (who held a high proportion of management positions prior to 1972) by Malagasy staff, whose training and degree of preparation have varied widely. Finally, Madagascar's effort to decentralize local government and administration through the Fokonolona system, a four-tier hierarchy of representative institutions based on tradi- tional democratic village government, has figured prominently in the Govern- ment's stated goals for rural development, and the new institutions have begun to play an active day-to-day role in all local affairs. B. THE AGRICULTURAL SECTOR - BACKGROUND 4. Agriculture dominates the economy of Madagascar, and about 85% of the population of about nine million live in the rural areas. Agriculture contributes about 40% of total GDP, accounts for more than 80% of export earnings, and supports directly over 80% of the population. The growth of agricultural production in the past 15 years has been disappointing, with total production increasing on average at less than 3% a year. This perfor- mance largely reflects the low rate of growth of rice production, and the stagnation or even decline of the national cattle herd. Production of some crops has increased significantly, notably cotton and coffee, but because of the paramount importance of rice and livestock production, their impact on the overall economic picture has been limited. 5. Madagascar's agricultural sector is striking in its diversity. The rural economy varies among several sharply different geographic regions; Madagascar produces a very wide range of crops and other agricultural products, and a variety of production systems are employed, ranging from large-scale mechanized and sophisticated production systems to the simplest, small-scale traditional farming. This diversity has to some extent protected Madagascar against the vicissitudes of international commodity prices and adverse cli- mate (drought and cyclones being ever-present hazards). It also contributes to the wide disparity in wealth and level of development among geographic regions which characterizes Madagascar; the most dramatic contrast is between the central plateau area and particularly Antananarivo, and the coastal areas of the west and the south.  - 26 - ANNEX 1 APPENDIX Page 3 Geographic Regions 6. Geography has had a vital influence on Madagascar's agricultural development. Climate, topography, soils, and natural vegetation vary widely among' different regions of the island, and have influenced the development of quite distinct production systems. In addition, communication links in many parts of Madagascar -are very difficult, and the road network is in gen- eral poorly developed; this has contributed to regional distinctiveness and to the isolation of many areas.' Rice cultivation and cattle husbandry are two common denominators found in almost every part bf Madagascar, but mo'st other'agricultural activities are-concentrated in a few specific areas. There are five' majoi agrcicultural regions (see map): (a) the-central plateau includes the highlands extending north - south in'the center of the island; the plateau is hilly, bisected by small- valle'ys and a few larger basins, -and climate ranges from sub-tropical to-temperate. Population is concentrated in this region, and most suitable land i's unde'r intensive cultivation.- Rice is the dominant crop, supplemented by other food crops, fruits and vegetables, and cattle; (b) the'east coa'st constitutes a long, narrow belt; climate is tropical, rainfall is high, and. the cyclones which strike Madagascar annually concentrate their force in this area. Population here is relatively dense; coffee is the dominant cash crop', with r.ice the staple food crop; (c) the northern zone also has a tropical climate and high rainfall, but topography is more varied than on the east coast. Population is less dense, and is concen- tra'ted in the major production zones. Many of Madagascar's export and indus- trial crops are concentrated here, notably'spices (vanilla, cloves, pep'per) and suga'r; (d) the west coast constitutes a-vast plain intersected by broad river valleys. Climate is drier, although rainfed cultivation is possible in most areas. The region offers ex'cellent natural pastures'while the valleys have rich potential for irrigated cr6ps oir cultivation on the flood plains, but most areas remain relatively underdeveloped and population is sparse; (e) the south is Madagascar's least developed region; climate is- relatively arid and agricultural'potential is" limited also by soils and topography. The population depends largely on traditional cattle'rearing for their livelihood. These regions do not coincide even'roughly with adminis'trative boundaries in most cases; the plateau region includes Tananarive Faritany, most of Fianarant- soa, and part of Tamatave; the south includes- most of Tulear Faritany and Diego Suarez is in-the north; the'west coast includes'Majunga and part of Tulear, and the east coast includes parts of Tamatave and Fianarantsoa faritanys (see Map). Thus, most' faritany's include a range of climatic and topographic conditions. Major Agricultural Activities 7. Madagascar produces an extraordinarily wide range of agricultural products; it is in fact quite difficult to find crops, fruits or other spe- cies which do not flourish somewhere' in Madagascar. Rice, however,-occupies the prominent position in the agricultural sector; it is grown on almost one half of all cultivated land, and is the staple food of most of the popu- lation. Cattle also play a vital role: long the traditional symbol of Madagascar, cattle are reared throughout the country, while beef is regarded as a supplement to rice in the staple diet of the people. The most important  - 27 - ANNEX 1 APPENDIX Page 4 cash crops (which are also the largest contributors to exports) are coffee, cloves, vanilla and pepper. Major industrial crops include sugar and cotton. Agricultural activities are summarized very briefly below with a focus on particular development problems; data on area cultivated and production, marketed output, and exports are presented in Tables 2, 3 and 4. (a) Major Foodcrops. Rice: an estimated one million hectares are under rice; total production is estimated at between 2 and 2.2 million tons a year, and the average yield about 1.8 - 1.9 tons per hectare. Between 80 and 90% of production is consumed on the farm. Of land under rice, about 75% are family rice fields, and the remaining 25% irrigation systems which have been developed by the Government or received Government support at some time. The Government has tried to promote rainfed rice cultivation on hill tops, but this represents only about 2% of cultivated area. Rice is produced throughout Madagascar, but is concentrated in the high plateau area and in the northwest. Data on rice production is poor, and trends in production are difficult to discern or to analyze with accuracy. It appears, however, that overall pro- duction is increasing slowly, by somewhat less than the population growth rate, while marketed production has stagnated or even declined in the past eight to ten years. Because of. the dominance of family rice fields, which rely on partial water control, climatic variations have a major influence on production and total production fluctuates significantly; in 1977/78 the crop was expected to be very poor because the rains were too late and too little. The result of these developments is that Madagascar has changed from being a net exporter of rice to relying heavily on imports; rice imports represented 7% of the total value of imports in 1977. Rice imports have evolved as follows (in tons): 1/ 1973 1974 1975 1976 1977 27,758 130,134 63,643 71,243 105,190 The most urgent requirements for increasing rice production would appear to be improvement of irrigation and water control systems, introduction of new high-yielding varieties, a reorganization of basic extension services, and revision of pricing policies. Manioc is the second important food crop, and most production is consumed on the farm. Other food crops are maize, beans, sweet potatoes, and potatoes, but these are less important in terms of total output. (b) Major Export Crops: Coffee is grown mainly on the east coast, largely by smallholders, production has been increasing in the past few years, largely as the result of a program to rejuvenate coffee plantations and in- troduce new techniques. Production was estimated at about 79,000 tons in 1976 and 89,000 tons in 1977; as a result of the recent boom in coffee prices, coffee contributed almost 50% of Madagascar's total export earnings in 1977. Cloves, Vanilla and Pepper occupy relatively small land areas in specific 1/ Central Bank estimates, April 1978.  - 28 - ANNEX 1 APPENDIX Page 5 geographic areas, but are of major importance for exports. Madagascar is a leading world producer of cloves; production is concentrated on the east coast, north of Tamatave. Production has increased steadily in the past decade, despite variable production reflecting a normal four-year cycle; production in 1977 was estimated at about 15,000 tons. Vanilla production is also concentrated in the northeast, and Madagascar is the world's leading vanilla producer. Most is grown by smallholders, by an estimated 38,000 vanilla growers, and production is about 5,000 tons (green vanilla) a year. Pepper also is produced in the northeast and west; production ranges between 3,000 and 5,000 tons a year.- (c) Industrial Crops. Sugar is produced in four existing estates, one on the east coast, and three in the northwest. Sugar production has remained relatively constant, and new investments have not been made for some time. About two-thirds of sugar production is consumed domestically and the re- mainder is exported, most to the EEC. Cotton production has increased dramatically in the past decade. It is produced on the west coast, mostly on flood plains as floods recede, but also under rainfed conditions and under full irrigation. Production was estimated at 37,000 tons in 1977. About 51% of production comes from private expatriate companies, 34% from smallholders, and 15% from state farms. CFDT has had overall responsibility for seed mul- tiplication, marketing, and ginning, and has generally supervised provision of inputs, farmer extension, and management of state farms; its responsibil- ities are being assumed by a Government-owned company. Cotton is used for the local textile industry, and Madagascar now exports cotton manufactures. Tobacco production was formerly an important cash crop, particularly in the current cotton producing areas; however, production has declined recently, largely because of competition from cotton. (d) Other Crops. Oil seed production has increased slowly in recent years, and Madagascar imports increasing quantities of fats and oils. Oil palm and coconut schemes are underway on the east coast, and in the west efforts are being made to promote peanut production, which has declined in recent years. The inefficiency of the processing industry, limited exten- sion services, and controls on prices have curtailed growth. Madagascar produces a wide range of minor commodities, including wine, tung oil, medic- inal plants, ylang ylang, raffia and aleurites; none make a major contribu- tion to total output, but they have important local effects and collectively could make a significant impact on total agricultural production and exports. (e) Livestock. Cattle are the traditional symbol of Madagascar (the curved horns of the Malagasy zebu are pervasive emblems of the nation), and much of the population is heavily dependent on livestock production. Madagascar also has important natural advantages for livestock production, with a comparatively good position regarding animal diseases and vast areas of natural pasture. The cattle herd, variously estimated at between 7 and 11 million head, is virtually all raised on an extensive basis, owned by tradi- tional herders who use virtually no modern techniques and make almost no cash investments for livestock. Herd productivity is low and only a small part of the herd enters commercial channels (perhaps 3-4% of the total herd a year). Madagascar has traditionally exported beef, primarily to France and to other Indian Ocean islands. Together with other beef-exporting ACP countries  -29 - ANNEX I APPENDIX Page 6 (Kenya, Botswana, and Swaziland). Madagascar obtained a quota for duty free exports to the EEC under the Lome Convention. A long-term program for devel v- ment has focused on construction of export slaughterhouses, and two large new plants are now operating (Tananarive and Morondava), and a third was scheduled to be opened in Majunga in June, 1978. The problem is that production is stagnating, and marketed supplies are probably declining, while domestic demand has rapidly increased. It is also postulated that the size of the national herd is declining, although no clear reason for such a decline is apparent and data is so poor that definitive analysis is not possible. The principal causes are lack of a clear program to aid traditional livestock producers, restrictive pricing policies which limit financial incentives to producers, and a marketing policy, aimed to reduce the role of private inter- mediaries, which has yet to produce significant results. There is excellent potential for other forms of livestock production in parts of Madagascar, notably pigs, poultry and dairy; the Middle West region, west of Tananarive, in particular offers good potential for mixed farming. (f) Forestry production is of major importance in two areas, along the east coast escarpment east of Tananarive, and in the southern plateau area around Fianarantsoa. Here about 100,000 hectares have been reforested with pine. Major projects based on forest industries are currently under study. (g) Fishing is increasing in importance; Japanese fleets fish Malagasy coastal waters, and have been largely responsible for a marked increase in shellfish exports. Otherwise fishing, both sea fishing and inland fishing, offers excellent potential, but it has not yet been exploited. Farming and Production Systems 8. Agricultural production systems vary from sophisticated, mechanized production systems, as for cotton in the northwest, to very simple subsistence agricultural techniques. Private enterprise has dominated agricultural pro- duction in the past, but the Government is playing an increasingly active role; most noteworthy is the nationalization of the sugar industry in December, 1976, and the Government's assumption of responsibility for marketing of rice and other foodcrops, for cattle marketing, and for meat processing and exports. The Malagasy government is emphasizing the role of parastatal enterprises in its new political orientation ("socialist enterprises"). State farms under the Ministry of Rural Development also are involved in a range of crops. The Government is now emphasizing socialist cooperatives for farmers, but very few exist to date; the same applies for collective production within the fokonolona system. Traditional, largely subsistence farming dominates Madagascar's agriculture despite these developments, and most of Madagascar's estimated 1.5 million farm families are smallholders; a large proportion of families own some cattle. Traditional land rights prevail in most of Madagascar; these recognize individual rights to cultivated land, with rights passed on to heirs, and collective (fokonolona) rights to pasture. Farming techniques for smallholders vary quite widely among regions, ranging from relatively sophisticated rice cultivation techniques involving elaborate terracing and partial water control systems to slash and burn cultivation and semi-nomadic cattle husbandry.  -30 - ANNEX I APPENDIX Page 7 Institutions Responsible for Agricultural Production and Development 9. A large number of institutions have responsibilities for agricul- tural production and development, and their organization reflects both the complexity of the sector as a whole, and the historical evolution of insti- tutions and production patterns. The major institutions and their current responsibilities are as follows: (a) The Ministry of Rural Development and Agrarian Reform is the central institution for rural development programs. The Ministry has four major operating departments, which operate quite autonomously; they are organized separately at a regional level, normally down to the fivondronan- pokonolona (sub-prefecture). The Agriculture Department is the "classic" extension service, concerned principally with plant protection and to some extent extension; Rural Engineering (Genie Rural) develops and operates official irrigation systems, provides other services such as tractor hire and water developments, and oversees Ministry construction projects; the Livestock Department is principally concerned with provision of basic veterinary ser- vices, and is also responsible for marine fisheries; and forestry is respon- sible for reforestation programs, but also seeks to control brush fires (a widespread problem), and to promote inland fisheries. A fifth Department of Programming is responsible for planning, and for overseeing semi-autonomous regional authorities, which are part of the Ministry itself, and parastatal agencies. The Department of Agrarian Reform is principally concerned with redistribution of land formally farmed by expatriates. (b) Regional agricultural extension programs: In the plateau region, a number of relatively intensive extension services have been developed with assistance from external donors (FED and FAC); most of these organizations continue to function as relatively autonomous entities. Most were part of the GOPR (Groupement d'Operation des Productions Rizicoles), which was sub- divided by regions; the extension organizations include the URER's (Unite regionale d'expansion rurale) of Itasy, Tananarive, Antsirabe, Ambositra, and Fianarantsoa, the ZER (Zone d'expansion rurale) of Ambilobe, and ODEMO, a development agency for the Middle West. An on-going UNDP rural development program in the south is developing its own basic organization and services (Operation Androy) (Table 1). (c) About 26 parastatal agencies operate under the aegis of the Ministry of Rural Development (Table 1); some are regional development authorities, normally created in conjunction with externally-funded development projects (5 of these financed by the Bank); others are more akin to state farms organ- ized for specific production activities (silk culture, coffee, citrus, and ranching, for example). (d) Promotion of the principal export crops is principally organized through stabilization funds: these finance extension services, road develop- ment, and crop protection (among other activities) from their own resources; services are generally supervised by the Ministry of Economy and Trade at present.  - 31 - ANNEX 1 APPENDIX Page 8 (e) Agriculture credit is the responsibility of the BTM 1/ - rural credit bank. Although the BTM was established only in December 1976, it succeeded directly the agricultural services of the BNM-Malagasy Development Bank, a well-established and sound development institution. The BTM operates quite autonomously; its principal links with the Government are through the Ministry of Finance and Plan. (f) Agricultural Research has been centralized since 1972; previously, research was carried out by a large number of semi-autonomous agencies and institutions, most managed by external (largely French) institutions. All research now comes under the Ministry of Higher Education and Scientific Research; CENRADERU (Center for Agricultural Research for Rural Development) or FOFIFA, as it is also known, is a semi-autonomous body responsible for all agricultural research (including socio-economic analysis); it has few formal links with extension services, as it comes under the central research authority. (g) Marketing operations come under various parastatal companies or agencies, generally guided by the Ministry of Finance and Plan or of Economy. The Central Bank has significant responsibilities now in the area of rice marketing. (h) Local government institutions (the fokonolona system) have assumed important responsibilities, notably for marketing, but also for planning and implementation of development projects. The hierarchy of fokonolonas com- prises the fokonolonas or fokantanys which coincide roughly with traditional village units (about 11,000), the firaissam-pokonolona, at the canton level (about 1,100), the fivondronan pokonolona, at the sub-prefecture level (about 100), and six faritanys, which have replaced the former provinces. The Ministries of Interior and of Information and Ideology have primary respon- sibility for supporting the fokonolona institutions. The general result of the plethora of institutions involved in rural devel- opment is a fragmentation of authority for development programs, both on a regional basis, because of the predominance of regional authorities which enjoy considerable autonomy, and at a central level, because of the division of responsibilities among a series of ministries and agencies. C. GOVERNMENT STRATEGY AND PLANS FOR RURAL DEVELOPMENT 10. The Government of Madagascar's broad objectives for rural develop- ment are articulated in a new development plan, approved by the National Assembly in November-December 1977. The plan includes a long-term strategy, for the period 1978-2000, a three-year plan for 1978-80, and a program for 1978. The new plan does not represent an abrupt departure from the previous 1974-77 plan, but is much more expansive in its treatment of socialist insti- tutions and objectives. In essence, Madagascar's stated objectives for the year 2000 are to transform the society and to develop a strong and self- sufficient basis for development; the period 1978-84 is seen as a period of 1/ BTM is the Malagasy acronym for Rural Development Bank.  - 32 - ANNEX 1 APPENDIX Page 9 laying the foundations, 1985-92 are for consolidation, while 1993-2000 are to be years of expansion and growth. In the long-term plan, three goals are emphasized: (a) attainment of self-sufficiency in food and development of basic industries, largely supplied by the agricultural sector; (b) decentral- ization of all economic activities, through the fokonolona system, and assump- tion of a more active and direct Government role in the economy, and (c) narrowing the gap in distribution of incomes, and promoting more balanced growth among regions. The plan states that, particularly in the early phases of the plan, agriculture will represent the basis of all developments, and must receive a major share of investment and effort. 11. The plan for 1978-80 sets out more specific objectives for agricul- ture. Agriculture is expected to grow more slowly than other sectors: aver- age growth rate of value added is expected to be 4.3%, while the rates for mines and energy and industry are estimated at 8.8% and 10.7% respectively. Value added of processed agricultural products like sugar, rice, fats, tex- tiles and others are forecast to grow at higher average rates than the sector as a whole. In 1980 individual farms are expected to contribute about 80% to value added generated in agriculture, cooperatives 10%, socialist enter- prises 2.5% and private enterprises 2.5%. Investment figures are not clearly projected in the plan; it seems, however, that only 15% of total investments in real terms are assumed to be allocated to agriculture, most of it by the Government. The plan states that the share of agriculture in productive in- vestments is 27.7%, without, however, defining the term productive investments. The plan recognizes that, in spite of the growth rates assumed for production, employment possibilities in agriculture will increase very little, reaching about 3.9 million in 1980 (versus 3.5 million in 1975). Industrialization is considered the only feasible long-term solution for the problem of unem- ployment. The plan aims to improve rural living conditions, thereby limiting migration to the cities. Internal, rural to rural migration is emphasized, principally to the faritanys of Diego Suarez (Antsironana), Majunga and Tulear. In general, it is difficult to interpret plan targets for agricul- ture because the base data used is unclear and internally inconsistent. 12. The plan emphasizes the linkage of Madagascar's economical, poli- tical, social and cultural development to the establishment of a socialist regime. Decentralization consisting of the creation of socialist units (the fokonolona institutions), and people's councils and executive committees is the focal point of the organization of revolutionary power. For agricul- ture, the major proposed changes entail the introduction of an agrarian revolution and the creation of socialist cooperatives; the agrarian revolu- tion principally involves Government assumption of control of cultivated land, particularly estates or plantations which have been lying idle .for some time, or newly-developed land; socialist cooperatives would theoretically be respon- sible for this land. The plan does not elaborate on the status of implementa- tion and the amount of transformation envisaged during the plan period. The plan foresees state control over basic industries, particularly those related to agriculture. Though again the goals for the plan period are imprecise, the plan states that at a later stage the fokonolonas and the socialist coop- eratives should manage these basic industries. As regards marketing and  - 33- ANNEX 1 APPENDIX Page 10 distribution of goods, the Government seeks to extend control over the whole sector, including internal and foreign trade; the plan predicts a more effi- cient system once cooperatives have been established. The plan emphasizes the need to develop agricultural services but does not specify how. Regard- ing future agricultural projects and programs, the plan focuses on general objectives and targets; however, three themes are of key operational signi- ficance: (a) Rehabilitation of existing irrigation systems is emphasized rather than development of new systems; (b) the emphasis in development pro- grams is to be on projects which benefit large numbers of farmers at a low cost per beneficiary; and (c) despite major socialist transformation, the plan acknowledges that producer prices remain the most efficient way of motivating farmers. D. ON-GOING DEVELOPMENT PLANS AND PROGRAMS 13. Madagascar's total government budget for 1978 allocates FMG 87,992 million (US$383 million) for the recurrent budget and FMG 36,374 million (US$162 million) for the capital budget; this does not include programs funded by external grants or loans (except counterpart funds) nor various autonomous budgets. Of the recurrent budget, about 8% is allocated to the Ministry of Rural Development. Of the investment budget, the Ministry's share is about 15% of total investments. The operating budget lists about 65 separate rural development operations or projects; most are relatively small in scale, including a large number of pilot schemes and support of state farms. The most substantial investments financed by the government budget are the development of civil works, assistance to SODEMO, and development of sugarcane, all in the Morondava area, the broadly-defined "rice catch-up program" (Ratrappage paddy), provision of equipment and a mechanization workshop, a peanut development program, an irrigation system rehabilitation program (financed in part by the AfDB), two forestry projects, and village livestock development in Majunga Province. 14. The major externally-financed development schemes for agriculture and rural development may be summarized as follows. The European Development Fund (FED) is financing a substantial development program in the Andapa basin (north-east), a "micro-hydraulique" program for small-scale water control investments, a "micro-realisations" program for small rural works, such as storage facilities, a tea program, development of irrigation systems in Marovoay and the Mangoky delta, and several studies for future projects. The French government continues to provide substantial technical assistance, and is financing construction of a new slaughterhouse at Majunga, and ex- tensive studies for development of the Lac Alaotra basin. UNDP's current role in rural development is limited, and is focused on a rural development program in the Androy region in the south. The Government has recently requested UNDP assistance for agricultural research. The African Development Bank is financing rehabilitation of three medium-sized irrigation systems. Other bilateral programs include, inter alia, Norwegian assistance for live- stock on the plateau, south, around Fianarantsoa, Swiss assistance for dairy development, Chinese assistance for a sugar mill (in Morondava), Russian assistance for studies of a major river basin (Mandrare) in the south, and North Korean assistance for various agricultural developments.  - 34 - ANNEX 1 APPENDIX Page 11 15. The Bank Group has financed five projects for agriculture in the past: (a) Beef Cattle Development Pro'ject (Loan 585-MAG) - completed (b) Lac Alaotra Irrigation Project (Credit 325-MAG) - completed (c) Morondava Irrigation and Rural Development Project (Credit 322-MAG) - ongoing (d) Village Livestock and Rural DIevelopment Project (Credit 506-MAG) - ongoing (e) Mangoro Forestry Project (Loan 1065-MAG, Credit 525-MAG) - ongoing The Mangoky Agricultural Development Project was appraised in April 1978. Two additional projects have been identified and are being prepared - an agricultural credit project, and a "fokonolona" rural'development project. IFC has financed the expansion of textile mills in Majunga, and is consid- ering an investment in the sugar industry. An education project currently being prepared includes an agricultural education component. 16. Experience with past projects has been mixed. In summary, the Beef Cattle Development Project-and the Morondava Project have encountered very serious problems. The Forestry Project and the Lac Alaotra projects have been,generally quite successful. The Village Livestock project encoun- tered seriods problems at first, but now appears to be going quite well. E. CENTRAL PROBLEMS FOR AGRICULTURAL DEVELOPMENT 17. Madagascar faces many difficulties in its efforts'to reorganize the agricultural sector and the rural economy, and to stimulate agricultural production. Recent performance of-the agricultural sector has been dis- appointing, albeit with nxotable-exceptions of promising growth in selected regions or crops. The reasons for agriculture's sluggish growth are complex, and there is much concerning farmer response to various incentives and the behavior of the subsistence economy which is not well understood. Several issues, however, have emerged as recurrent themes both in general economic discussions between the bovernment and the Bank, and in project operational work', and should serve as a focus of future discussions of policy and strategy, and of future project preparation work: (a) Pricing and marketing policies; (b) Institutional problems including weakness of extension services-and input supplies to farmers; Cc) Limited planning capacity for agriculture and poor pro- cedures for preparing and evaluating project proposals;  - 35 - ANNEX I APPENDIX Page 12 (d) Poor supervision of ongoing development programs and weak coordination among donor agencies. Another central problem for agricultural development is the poor road network which impedes communications throughout Madagascar and complicates both plan- ning and implementation of development program. Improvement of both the basic road system and feeder road networks and improved road maintenance are of critical importance for rural development efforts. Training of staff for all institutions is another continuing problem; this includes training of techni- cians and extension workers, management training now that Malagasy nationals are assuming key management positions with relatively limited preparation, and training of staff of fokonolona institutions in all fields, including management, finance and administration, planning, and basic technical skills. Pricing and Marketing Policies 18. Government pricing policies and efforts to reform and control marketing represent an important bottleneck for development of several key sub-sectors of agriculture, notably rice, livestock, oil crops, and sugar. Other crops are affected to a lesser extent. The problems result from two very reasonable Government objectives: to limit inflation and particularly the cost of living for the urban poor, and to improve the efficiency and reduce the cost of marketing. Problems have arisen because producer prices are often fixed at too low a level, while changes in marketing have been instituted without adequate planning or evaluation of the consequences. In the past, the urban consumer has tended to benefit from pricing policies at the expense of the rural producer, thus exacerbating disparities between urban and rural incomes. For rice, an additional problem is the Government's decision to reduce prices to consumers and to pay a direct consumer subsidy. The key problems may be summarized as follows: (a) Rice. The Government intervenes at four stages of marketing and pricing of rice: (i) producer prices for paddy are fixed by the Central Gov- ernment, with the final decision taken by the Cabinet or the President himself, normally on an annual basis; (ii) domestic rice marketing is a state monopoly, and the Government controls all internal marketing arrangements; (iii) rice import requirements are estimated by the Government, which also controls imports and distribution of imported rice; and (iv) consumer prices of rice are fixed by the Central Government, again on an annual basis; the current consumer price entails a direct subsidy of about 40% for locally produced rice 1/ and a larger subsidy for imported rice. There is a significant black market in rice, and the market price is well above official levels. Rice pricing and marketing raise three principal issues: 1/ Current prices are as follows: producer price of paddy FMG 35/kg; marketing commission of fokonolona FMG 2; wholesale price of rice FMG 53.5/ kg, retail price of rice FMG 55/kg. The real cost of rice is estimated at FMG 70-75/kg. The 1977 average cost of imported rice (CIF Madagascar port) was about US$325 per ton or equivalent to about FMG 73/kg. Costs have increased sharply in 1978, and the import price was estimated at FMG 90-95/kg, CIF Madagascar port in April 1978.  - 36 - ANNEX I APPENDIX Page 13 (i) Adequacy of producer prices as a stimulus to production; prices are currently below import parity levels, and an increase in price might well result in production increases; (ii) Efficiency of Marketing: responsibility for rice mar- keting is now divided among 12 parastatal companies, working in support of fokonolonas who have primary responsibility for marketing within their territorial jurisdiction. Marketing has not operated efficiently in the past; not all rice is collected, regional short- ages occur frequently, and marketing costs are high. A systematic review of marketing organization should be viewed as a high priority for the Government; and (iii) Consumer prices and consumer subsidies: the Govern- ment now subsidizes consumer prices, and this issue should be evaluated in light of the impact of sub- sidies on the Government fiscal situation and the' Government's social policies for assisting the urban poor (who are the principal beneficiaries). Subsid- iary issues include the need to determine the precise need for subsidies (i.e. the true cost'of marketing) and to arrange for prompt reimbursement of such sums to marketing agencies. (b) Cattle Marketing and Meat Prices: Meat prices are controlled by regional authorities (faritany level), and have been set at levels well below the export price equivalent. Medium quality meat (bone in) sells for about FMG 260 per kg (US$1.16) and the price has remained virtually unchanged for about 3 years. No premium is paid for differences in meat quality. The market price of meat is 25-50% above official prices. One firm obtained an average of FMG 434/kg net for exports of frozen beef. A high proportion of beef is reportedly sold on the unofficial market, at prices well above offi- cial levels. Prices of pork, which are subject to less stringent controls, are now double those of beef. Price controls have resulted in an increase in beef consumption within Madagascar, a decrease in meat exports (Madagascar has a duty-free quota of about 10,000 tons a year on the EEC market but does not fill this quota). Further, price controls have discouraged investments in livestock development and have contributed to decreases in marketed supplies of cattle. The Government is also moving to nationalize cattle marketing and to control producer prices of cattle. Parastatal agencies have been assigned responsibility for cattle marketing, and these agencies are also assuming responsibility for meat exports, taking over from private firms which have gone out of business or have been nationalized. The official margins between producer prices of cattle and meat prices have been too small to permit devel- opment of efficient marketing systems, and the Government subsidizes the marketing agencies directly, although without an explicit policy to this effect. Livestock pricing and marketing raises serious issues, but more information is required before alternative strategies and policies can be discussed; a major study by the Government assisted by SEDES, financed under Credit 506-MAG, is being undertaken. Discussions of policy options and issues should be held at the conclusion of this study, in late 1979.  - 37 ANNEX I APPENDIX Page 14 (c) Other Products: Pricing policies for other key commodities raise general issues of strategy. They should be reviewed in conjunction with the economic updating mission. Institutional Problems 19. Both the Madagascar government and the Bank recognize the need to accord a high priority to development of effective institutions for agricul- ture and rural development. The most immediate problems affecting institu- tions may be summarized as follows: (a) Responsibilities for planning and implementation of agricultural programs is fragmented among a large number of institutions and agencies, including both Central Government and regional authorities. Taking rice as an example, different agencies are involved, respectively, for basic and applied research, extension, irrigation, provision of inputs, marketing, pricing, and imports. The Ministry of Finance and Plan has responsibility for coordination of planning, but lacks the staff to prepare project directly to follow all preparative activities closely, and to evaluate on-going and completed projects; the same applies to the Ministry of Rural Development's programming department. The development of autonomous authorities and agencies to implement development programs has been particularly marked in Madagascar, and many officials still believe firmly that a separate statutory authority is vital if programs are to be carried out effectively. However, creation of these authorities has left two legacies: first, agencies created to satisfy different donor agencies vary widely in character, procedures, and structure and there is virtually no unity in approach; and second, as donor funds cease to be available, agencies are often left without clear objectives and with inadequate funding. In general, a careful review of existing institutional arrangements and probably a substantial reorganization would appear essential prerequisites for launching a coherent and efficient agricultural development program. (b) The Government is seeking to decentralize Government services, notably for agriculture, to the fokonolona local government institutions. Considerable uncertainty exists at present as to the extent of local Govern- ment responsibilities, their financial means, and the respective authority of central ministry staff and local officials. Further, training of local government staff, notably the vatoekas (economic committees) of fokantanys, should receive immediate emphasis. (c) Uncertainty about the pace of change to socialist institutions and the specific objectives of the Government for socialist enterprises and socialist cooperatives clouds the current institutional picture. There has been talk of major reorganization of the Ministry of Rural Development for some time but no formal decision has yet been announced. 1/ This general aura of uncertainty itself reduces the effectiveness of many officials and contri- butes to delays in projects as people await the outcome of expected reforms. 1/ This reform is currently underway following a decree of April, 1979.  - 38 - ANNEX I APPENDIX Page 15 (d) Madagascar has no basic agricultural extension service, and current links between research and extension are weak. Extension services are prin- cipally'provided by a'large number of autonomous agencies and parastatals, organized on an ad hoc basis in the past. The development of a basic exten- sion program for key crops and establishment of working links between research and extension merit a high priority. Supplies of inputs to farmers, including in particular seeds and fertilizers and credit are an important bottleneck to increased agricultural production. (e) There is a lack of coordination between the principal agencies responsible for project preparation and for their implementation; this has resulted in serious difficulties during the implementation stages. This is the case for many ptojects in Madagascar; for example, no research on peat soils in the Lake Alaotra Project; limited technical recommendations on varieties of crops to be grown; no extension program in harmony with crop planning in'many agricultural development projects; and incomplete pricing and marketing policies' for imported-agricultural 'inputs and outputs in all projects. There is an urgent need for more systematic procedures for coordination. Agricultural Planning and Project Preparation 20. The Department of Planning in the Ministry of Finance and Plan is responsible for all planning, but its staff and resources are limited. The Ministry of Rural' Development also lacks staff and resources for general planning. At piesent no'clearly defined procedures exist for preparation of projects. Development of proposals for specific donors, tailored to their needs and requirements, has been the general rule in the past, and most projects have been prepared by expatriate consulting firms, on a rather ad. hoc basis. In many'cases studies by these firms have resulted in voluminous documentation poorly. adapted to Madagascar's priorities and resources; the- piles of documents tend to languish in offices and are soon outdated and virtually useless. There is an'urgent need to clarify basic procedures for project preparation and for appraisal of project proposals, beginning with decisions on allocation of basic responsibility followed by a review of the costs and benefits of alternative methods of project preparation. Supervision of On-going Development Programs and Coordination of Donor Activities 21. The Government has difficulty in supervising the multitude of on- going development projects, programs, and institutions. The case of SODEMO (Credit 322-MAG) which produced no accounts over four years of operations and ran substantial deficits beyond planned levels without intervention by super- visory ministries illustrates well the difficulty of overseeing the numerous agencies which have developed. The need to review basic procedures for financial and administrative control, and for evaluating-the effectiveness of on-going operations is urgent. There is also a need to coordinate more effectively the development programs of various donor agencies to ensure that all fit with Madagascar's basic goals for development institutions, and to minimize overlapping and duplication.  - 39 - ANNEX I APPENDIX Page 16 F. PROJECTS UNDER DISCUSSION OR REVIEW 22. Over the past two years, successive missions to Madagascar have discussed project possibilities in the agricultural and vural development sector; a number of promising activities have been identified by the Ministry of Rural Development and other ministries, some in conjunction with Bank staff members. Significant preparatory work has been undertaken, primarily by Malagasy officials, to assist in assessing the various possibilities. While the Ministry of Rural Development has not hesitated to assign priorities among projects and has aggressively sought Bank involvement, the Ministry of Finance and Plan has consistently reserved the Government's position on relative prior- ities, indicating that more time was needed for the Government to review its overall strategy for development of the rural sector. The status of prepara- tion of projects for possible Bank Group financing was discussed most recently during the Mangoky appraisal mission. The Director General of Plan presented the mission with the following list of project priorities for agriculture and rural development; this list was elaborated following a meeting to discuss these proposals including officials of the Ministry of Finance and Plan and the Ministry of Rural Development: Studies 1. Lac Alaotra 2. Livestock - Middle West 3. Livestock - Majunga Province 4. Citrus Orchard, Bezezika Projects to be Implemented 1. SAMANGOKY 2. "Fokonolona" Project 3. Plaine de Tananarive 4. Line of Credit, BTM 5. Morondava, Second Phase 6. Forestry II 7. Rehabilitation Petits Perimetres 8. Livestock - Majunga Province 9. Livestock - Middle West This ranking of priorities should offer a useful point of departure for future discussions; however, many of the proposed projects require considerably further definition. Further, the list includes only projects that have been under discussions for some time and offers little basis for significant new departures in formulating a coherent program to assist in Madagascar's rural development effort.  - 40- ANNEX 1 APPENDIX Page 17 G. PROPOSED STRATEGY FOR AGRICULTURAL DEVELOPMENT 23. In planning for Bank/IDA lending operations for agriculture for the next five years and beyond, the following steps and activities should be discussed and agreed upon: (a) Definition of a general strategy for Bank/IDA assistance for rural development; (b) Formulation of a general order of priorities among projects or types of projects; (c) Agreement on measures required to prepare these projects, including allocation of responsibility for preparation work, and estimation of costs and sources of financing; (d) Agreement on a specific timetable for project preparation work and a general lending program; (e) Continuing dialogue by both Programs and Projects staff on broad problems affecting rural development, notably pricing and marketing for major food products (rice, meat, oilseeds), institutional control of on-going development projects, and general organization of research, extension, and input sup- plies to farmers. General Strategy for Bank/IDA Assistance for Rural Development 24. The importance of the rural sector in Madagascar and its recent mediocre performance, and the limited number of operations supported by IDA in recent years suggest that we should make a concerted effort to assume more active role in the future. The focus of IDA efforts should be focused on specific project activities, complemented by general discussions and reviews aimed to assist the Government in designing policies for the sector as a whole. We should recognize that project work will raise difficulties, and a particular effort will be required to assist in or monitor preparation and to ensure an intensive supervision input. Given Madagascar's geographic diversity and the wide range of significant agricultural activities, it would be appro- priate to lend for a significant number of small projects affecting several regions and sub-sectors of activity, rather than to concentrate on larger projects or a single national rural development-program. However, we should also try through one or more projects to assist in strengthening central institutions and services. The total volume of funds available and the number of projects in the program must evidently be determined on a country and regional basis; however, we recommend that in the future agriculture should receive a significantly higher proportion of resources than it has in the past; given the magnitude of the need for resources and above all for coherent development schemes, we should consider whether a program including two or three agricultural projects a year might be feasible once a proper pipeline of projects has been developed. This might, of course, also offer significant prospects for co-financing with other development agencies.  - 41- ANNEX 1 APPENDIX Page 18 25. Order of Priorities Among Projects: This will require discussion with the Malagasy Government, and should take into account activities of other donor agencies. The objective is less to establish a ranked list of projects than to agree on allocation of projects by subsector as a basis for planning and preparation activities. 26. Project Preparation Methods and Costs: Alternative means of pre- paring projects should be reviewed, including (a) strengthening of prepara- tion units in the Ministry of Rural Development and the Ministry of Finance and Plan; (b) employment of consulting firms for specific preparation tasks, either local firms or internationally-recruited firms; (c) use of the FAO/CP to assist in preparation work; and (d) RMEA or IBRD/Washington direct assis- tance for project preparation. These are not mutually exclusive and a mix of preparation methods will be required. Malagasy officials have indicated that they may wish to establish a special project preparation unit or company, and these proposals merit discussion. A high priority should be accorded to direct assistance to the Government to strengthen its project preparation and appraisal capacity. This might include technical assistance and training. Some funds for preparation (about US$700,000) are available under on-going projects, but if requirements are greater than the available amount additional funds could be included as a component of the Mangoky Agricultural Development Project. 1/ 27. Timetable for Project Preparation Work: A general program of prep- aration work should be agreed upon, specifying allocation of responsibilities, financial requirements, and sources of funds. 28. Dialogue on Policy Issues: The June-July, 1978 economic mission will address the broad economic impact of agricultural development policies and problems. The review should focus on the impact of price controls and Government intervention in marketing. Continuing discussions on (a) the level of producer prices; (b) marketing arrangements for rice and other food crops and for livestock; and (c) subsidies to consumers will be required. When the SEDES study of meat prices and cattle marketing is completed (hopefully during 1979), general strategy for livestock development should be discussed fully with the Government. Government supervision of on-going projects and control of parastatal implementation agencies, general policies on research and extension, and an input supplies to farmers and project preparation and planning capacity merit continuing discussion; this may be most effective in the context of on-going supervision work but will also require high level discussions, notably in the course of project implementation reviews. Recommendations on Types of Projects and Categories of Activities 29. The following broad lines of activity are suggested as a basis for planning an agricultural lending strategy: 1/ The Mangoky Project, approved in FY1979, includes a project peparation component.  - 42 - ANNEX 1 APPENDIX Page 19 (a) Basic Services and Planning of Rural Development The principal objectives would be reinforce central services and contribute to development of more effective services to farmers. Four specific activities might be considered: 'i) Assistance to the Ministry of Rural Development for planning-and project preparation. Technical assistance, training, support services; (ii) Program to strengthen agricultural research, notably on rice; establishing stronger links between the Ministry of Rural Development and CENRADERU; assistance for applied research and trials; training of staff, including te-chnical assistance and training at inter- national research centers, such as IRRI; (iii) Assistance to the Ministry of Rural Development in developing a unified extension service; planning, training, and general support and improvement of input supplies to farmers; (iv) Assistance for strengthening local government institu- tions, training of staff of fokonolona institutions and assistance in planning for local project preparation activity and resource mobilization. These objectives will be furthered in conjunction with the next education project, which will include an agricultural training component. (b) Reinforcement of Specific Services for Sub-Sectors (i) Strengthening of Genie Rural, through a i"petits peri- metres" type project; the objective would be to improve productivity of existing irrigation systems and to establish clear and durable procedures for operation, maintenance, and financing of official irrigation perimeters. A general scheme of engineering assistance to family rice fields might be explored. This project should also include a general review of alternative techniques for irrigation development, including both a technical and financial evaluation. (ii) Development of effective veterinary and livestock extension services. Continuation of the on-going Village Livestock Project with possible expansion beyond Majunga Province, to other cattle breeding areas (notably Tulear). -(iii) A project to assist crop or livestock marketing services might be considered at some stage in the future.  - 43- ANNEX I APPENDIX Page 20 (c) Specific Crop or Production-Oriented Projects (i) The Agricultural Credit Project is likely to focus on directly productive activities, including export crops and miscellaneous cash crops (such as fruits, wine); (ii) Forestry Project. The proposed second forestry project would fall within this category; and (iii) Other Projects Related to Specific Crops - such as cotton and sugar might be identified in the future. (d) Regional Development Projects Several projects might best be viewed as regional development efforts; the focus might be a specific crop or sub-sector, or on land devel- opment and irrigation. These might include: (i) The Mangoky Agricultural Development Project: Land development, irrigation, and rice and cotton production in the south-west; (ii) The Middle-West Mixed Farming Project: A project to develop pig production, dairy, cattle fattening, and poultry, probably in association with crops in the Middle-West region, west of Tananarive; (iii) Lac Alaotra: Future project preparation work should emphasize the organization of production and provision of services for the region as a whole; development investment is likely to be dominated by irrigation and drainage infrastructure, but other facets of production and social services should receive immediate attention (distribution of inputs, livestock activities, etc.). (iv) Morondava II. If a future project for Morondava is prepared, it should have a regional development em- phasis. The central focus will be on maximizing returns from the existing irrigation system (for example, making the Bezezika state farm function), but development of livestock and dryland farming also offer promising possi- bilities; (v) Other potential regional development activities might be identified in Marovoay (rice producing region, south of Majunga), Bealanana/Ankaizina (northwest Madagascar) and other regions.  -44 - ANNEX 1 APPENDIX Page 21 (e) Plain of Tananarive: This proposed project raises special and rather complex problems. Agriculture is an important activity and productivity of rice could be increased, principally through introduction of new varieties. The problem of flood control is primarily a problem for the city of Tananarive, which has expanded (illegally) into areas highly vulnerable to flooding. Flood control would assist in bringing some new land (about 5,000 ha) into pro- duction, and in permitting a better control of water and hence higher yields and in some cases two crops a year; how- ever, the benefits would be small in relation to the likely cost of flood control. Further, flood control would stimulate urbanization of the current agricultural areas of the Plain, thus reducing agricultural productivity. If a development project is to be identified, its objectives must be clearly defined, and it must have an inter-disciplinary focus. 30. These project ideas and the current status of identification or preparation are summarized in Table 5. Evidently, all of these project ideas cannot be implemented, and decisions must be made on relative priori- ties. However, this summary of possibilities should serve as a basis of discussion both of the project preparation program and of the IDA/IBRD lending program for agriculture.  ANNEX 1 APPENDIX - 45 - Page 22 LIST OF PARASTATAL OR?'ANIZATIODIS UNDER THE JUISDICTION OF THE MIINISTRY Or RURAL DF7ELOP>MNT A. MIXED ECONOMY.COTPANIES (Soci6t6s d'Economie Mixte) - Societ6 Malgache pour l'Amenagement du Lac Alactra (SOMALAC) Malagasy Agency for the Development of Lac Alaotra - Bureau Central Laitier (B.C.L.) Central Milk Bureau - Societe"d'Etudes et de Developpement des Petimetres de Fiherenana de la Taheza (SEDEFITA or FIFATO) Study and Development Agency for Fiherenana and Taheza (Tulear Faritany) - Comite d'Expansion Economique de la Plaine de Marovoay (COME17A or FIFABE) Economic Develonment Committee for the Plain of Marovoay (Majunga Faritaany) - Societe pour l'Amenagement et la Mise en Valeur de la Vallee du Bas Mangoky (SAMANGOKY) Development and Production Agency for the Lower Z4angoky Valley - Soci6te pour le Developpement Economique de la Region de r4orondava (SODFAO) Agency for the Economic Development of the Morondava Region - Societe Malagasy pour le Palmier a Huile (SOM4APALM) Malagasy Palm Oil Agency - Fambolen-Kazon'Analabe Moramanga (FANALAMANGA) Moramanga Forestry Development Agency - Societe des Jus de Fruits d'Antsirabe (SOJUEA) Agency for Fruit Juices of Antsirabe - Fanjonoana Malagasy Koweitiana (FAMAKO) (Fishing-Ncsy be) B. OTHER PARASTATAL AGENCIES (Etablissements Publics a Caractere Industriel et Commercial) - Abattoir Frigorifique Industriel National de Tenanarive Tananarive Slaughterhouse - Org