Madagascar - Agricultural Strategy Note

Document of The World Bank for Official Use Only Report No. 10063-MAG MICROGRAPHICS CONFIDENTIAL Report No: 10063 MAG Type: SEC MADAGASCAR AGRICULTURAL STRATEGY NOTE February 23, 1994 Agriculture Operation Division South-Central and Indian Ocean Department Africa Region 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 autoization. CURRENCY EQUIVALENTS Currency Unit = Malagasy Franc (FMG) US$ 1.00 = FMG 1,925 (average 1992-93) SDR 1.00 = FMG 2,664 (average 1992-93) WEIGHTS AND MEASURES Metric System ABBREVIATIONS AND ACRONYMS DRC Domestic Resource Cost ESN Economic Strategy Note FAO Food and Agriculture Organization FOFIFA Agricultural Research Institute GDP Gross Domestic Product NGO Non Government organization OGL Open general License PSA Private Sector Assessment This report is a synthesis of extensive sfotor work Carried out during 1991-199. Benioit Blared coordinafted this work and prepared the first draft of this report. The present ver-ion was Pmpared by Adrian Otten and Mlichel Sirn&n. The Dimetor, the Economic Advisor and the Division Chief of the responsible departmient and division are P. Aguirre- Saoasa, U. Thurnun and N. O. Teheyan, resecdvely. ill TABLE OF CONTENTS EXECUTIVE SUMMARY INTRODUCTION 1 CHAPTER 1: PAST AGRICULTURAL DEVELOPMENTS 2 I. ROLE AND PLACE OF THE AGRICULTURAL SECTOR IN THE ECONOMY 2 H. THE CHANGING POLICY ENVIRONMENT 3 A. Policies Until the Early Eighties 3 B. The Reform Program (1984-90) 4 C. The Impact of Reforms on Agriculture and Exports 5 D. The Impact of Reforms on Rural Poverty and Food Security 7 CHAPTER 2: AGRICULTURAL POTENTIAL AND CONSTRAINTS 8 I. THE POTENTIAL OF AGRICULTURAL GROWTH 8 II. THE CONSTRAINTS TO AGRICULTURAL GROWTH 10 A. Inconsistent Macroeconomic and Sector Policies. 10 B. Lack of an adequate, market friendly legal and regulatory framework 12 C. Low Technology Use, Deforestation, Declining Soil Fertility and Land Degradation 13 D. Inefficient Farm Support Services, Degraded Rural Roads and Marketing Infrastructure, and an Unbalanced Public Expenditures Program (PEP). 14 CHAPTER 3: A STRATEGY FOR AGRICULTURAL GROWTH 18 I. THE VISION 18 II. THE POLICIES FOR GROWTH 19 A. Market Friendly Policies 19 1. Economy-Wide Policies 19 2. Sector Level Policies 20 B. Improving Farm Support Services 21 C. Improving Natural Resources Management 23 D. Infrastructure 24 E. Reshaping Public Expenditure 25 M. A SECTOR PROGRAM FOR ACTION 26 A. Preparation and Dialogue 26 B. The Proposed Approach 26 iv TABLE OF CONTENTS (continued) TABLES AND APPENDIXES TABLE 1: (A) National Accounts summary 30 (B) Annual Rates of Sector Growth 30 TABLE 2: (A) Volume of Agricultural Production 31 (B) Value of Agricultural Production 32 TABLE 3: (A) Importance of irrigation in Madagascar, by Province 33 (B) Importance of Paddy in Irrigation and in Total Cultivated area, by Province 33 TABLE 4: (A) Regional Differentiation in Agricultural Production (1989) 34 (B) Percent of Production 34 TABLE 5: Agricultural Exports (volume and value - 1984-91) 35 TABLE 6: Regional Economic Analysis of Agricultural Production in Madagascar (1991) 36 TABLE 7: National Income and Consumption 37 APPENDIX A: Regional Comparative Advantage and Pattern of Regional Specialization 38 I EXECUTIVE SUMMARY Past agricultural developments 1. Agriculture is the most important contributor to the national economy (34 percent of GDP in 1992). Agriculture employs over 70 per cent of the active population and accounts for 60-65 per cent of export earnings. The sector supplies the primary inputs to the food and textile industries which together account for 75 per cent of the value added of the industrial sector. 2. The agricultural sector grew at an average annual rate of 0.6 per cent between 1970 and 1980, well below the estimated population growth rate of 2.8 per cent. It declined at - 0.1 per cent on average between 1980 and 1983; with macro-economic policies concentrating on controlling inflation and the fiscal and balance-of-payments deficits, the provision of services and infrastructure continued to deteriorate. Trade liberalization measures were initiated in 1983 for rice, and generalized in 1988 (except for vanilla). Overall, the response from the agricultural sector to the adjustment measures throughout the 1980s has been mixed. It is only after 1988 that agriculture started responding to the new set of economic incentives; between 1983 and 1988, agriculture bottomed out, resuming growth in 1986, albeit below population growth (2.8 percent per annum). In 1989, agriculture grew by 7 percent, largely caused by a 12 percent increase in rice production; agricultural growth decelerated, however, in 1990 to 2.4 percent and again in 1991, estimated at 1.6 percent. 3. The supply response of agriculture varied across sub-sectors and regions. Fisheries and non- traditional exports, which benefitted from early trade liberalization and adjustment measures, significantly expanded after 1983. In the course of the last decade fisheries exports became the third higher foreign exchange earner. Similarly, non-rice agricultural activities located in the central highlands (dairy, potatoes, sweet potatoes, maize, barley, wheat) where the environment is more favorable (proximity of large urban markets, adequacy of physical infrastructure and agricultural services) also responded positiiely to improved price incentives. However, the traditional exports (coffee, vanilla, clove, pepper) and cotton sub-sectors, and their corresponding regions (East Coast, Northern region, West Coast) did not respond to the adjustment measures, to a large extent because of continued high levels of taxation during part or all of the period, coupled with a significant drop in world prices (coffee, cloves), and absence of full trade liberalization. The production of rice, after first taking off following the elimination of domestic price and trade controls in 1986, has also stagnated because of a poor market infrastructure (roads, markets and storage) and the absence of external trade liberalization and destabilizing government intervention. 4. Overall, the combined effect of domestic adjustment and deterioration of world market prices for traditional exports translates into a major shift in the structure of Madagascar's exports. Over 1985-91, traditlonal exports declined in aggregate value from US$284 million to US$95 million, and from 65 percent to under 40 percent of total exports. This reduction was partly compensated by an expansion of non-traditional exports, notably shrimp and fish, fruit and vegetables, textiles. Agricultural patential and Constraints 5. Potential. Madagascar has potential for accelerated agricultural growth. In terms of production, it has abundant arable land and diversity among ecological zones, enabling it to produce a variety of agricultural products ranging from tropical to temperate; it has extensive irrigation infrastructure which, if properly maintained, allows it to produce crops at high yields and with sufficient flexibility to meet changing market demand; and at present low yield levels for most crops, the possibilities for productivity improvement are huge. In terms of economics, it has a strong competitive base: present low labor costs allow it price advantages that enable it to substitute for 11 agricultural imports, and successfully compete on international markets. In terms of MAkA anilabilfty, Madagascar has - with the exceptions of vanilla and cloves - only a -dny share in the international market for its products. This is also true for regional markets (e.g., South Africa, Saudi Arabia, Mauritius, La Rdunion) where the country's proximity and easier trade links would give it an advantage over its competition. In general, market potential remains far from being fully exploited. In the short to medium term, an export-led growth scenario for agriculture would offer the best chances for rapid growth given the large diversity of products that it can provide at favorable prices in existing export markets. The strong growth and the growing diversity of non-traditional exports in the 1987-90 period give convincing evidence that such a scenario would be feasible. 6. Constraints. While Madagascar has a strong agricultural potential, it is also clear that it faces powerful and mutually reintorcng constraints to utilize this potential. The major growth constraints can be summarized under four groups: a) inconsistent macroeconomic and sector policies harmful to agricultural development, including: an overvalued exchange rate and the rationing of foreign exchange; price and/or trade distortions still remaining for vanilla as well as in areas where publicly or privately owned enterprises have de-facto monopolies in processing (cotton, sugar, wheat, copra, oil palm, beef for export); financial sector and credit policies that restrict access of small farmers to credit. Moreover, continuous declines in domestic per capita income, exacerbated by an increasingly skewed distribution of income, have restricted demand for agricultural produce in the home market; b) lack of a market-friendly legal and regulatory framework, resulting in excessive intervention by the public sector, the absence of development of a competitive private sector, and the lack of empowerment of local rural communities; c) low technology use leading to declining soil fertility and environmental degradation, including increased erosion due to improper hillside cultivation, expansion of the practice of bush-fires beyond the regenerative capacity of the plant cover, and accelerated forest destruction resulting from agricultural encroachment; and d) ineffective farmer support services and degraded rural roads and marketing infrastructure which do not allow markets to function efficiently, and a public investment program that is of limited help in addressing the above lim1w ions. A strategy for agricultural growth 7. Growth scenario. Only a comprehensive program of macroeconomic and sector policy reforms can bring about accelerated growth that would be sufficient to recover the 1971 level of per-capita income in about ten years, and to double it one generation. The high growth scenario presented in the Economic Strategy Note, with GDP growth reaching 6 percent per year by the year 2000, is based on the development by the private sector of diversified export activities with a high labor content. The six percent rate could be reached by combining a 4 per cent growth rate in agriculture with a 10 per cent rate in industry and a 6 per cent rate in services. 8. The structure of the Malagasy economy is such that a rural development policy based on the intensification of the small-farm sector in the high-potential, densely populated areas of Madagascar will accelerate the development of the entire rural non-farm economy, increase the size of the iI domestic market, as well as provide sizeable off-farm employment opportunities. A more aggressive export strategy represents a powerful instrument for boosting agricultural production and Incomes, but needs to be complemented with a strategy designed to satisfy the Increased local demand for food triggered by export expansion. 9. Main elements of strategy. To address the constraints that have impeded overall development In general and agriculture growth in particular, the Government would need to move on a broad front. First, a package of macroeconomic policies would have to be adopted, aiming at stability (with inflation, fiscal and balance-of-payment viability) and the restoration of growth, most Importantly through an open foreign trade regime and a competitive exchange rate. Second, a market friendly policy needs to be further pursued at the sector level, in which trade and prices are liberalized In the sub-sectors still under Government control (vanilla, sugar, cotton, wheat), markets are essentially left to the private sector, and foreign know-how and joint ventures with Malagasy firms are encouraged. A strong improvement in the efficiency of farm support services is needed, including a comprehensive decentralization to bring them closer to the field, both in management and decision making. For development to be sustainable, high priority is to be given to the proper management of natural resources. A large scale effort is required to improve rural infrastructure as well as farm to market links: better maintenance of irrigation structures and farm roads; rehabilitation of the rural and provincial road network; and better telecommunications. 10. Complementary measures. To effect these changes, a revision of the public expenditure program would require a shift away from current concentration on irrigation rehabilitation, towards better maintenance of rural infrastructure and improved operation of farm support services. A balanced public investment program would promote regional specialization and inter-regional trade, hence achieve a more efficient utilization of the country's resources. Increasing productivity and diversifying production on existing lands would do much to slow down the expansion of land use under low productivity techniques which is now occurring and which is the primary cause for environmental degradation. 11. Private agricultural institutions (e.g., professional and inter-professional organizations, farmers groups, water users associations, savings and loans associations, rural banks, input and commodity traders) would play an important role in developing market efficiency, increasing competition, and providing farmers with currently missing agricultural and marketing services, inputs and goods, and in overcoming market failures (e.g., information, quality, management and maintenance of shared infrastructure, credit rationing). Their emergence and expansion would be promoted by establishing an appropriate legal and regulatory framework and by providing for their support and training. 12. Implementation issues. The first step in this process is to arrive at a consensus on policies. At present, a forum for discussing such policies does not exist. It would be highly desirable to establish an agricultural policy unit within the Agricultural Ministry, with a mandate to (a) advise the Minister of Agriculture on the impact of economy-wide policies; and (b) within the overall policy framework, devise clear sector policies to achieve stated goals. An interministerial committee of high officials below the ministerial level should also be formed. It should include representatives of agricultural interests outside Government, such as trader and enterprise syndicates, farmers unions and NGOs. Once the group has reached a consensus on growth policies to be followed, the Public Expenditures Program for the sector should be subject to a comprehensive review in order to adjust it to the new priorities. At some point during this process, the Government should invite the donor community to support the new strategy through the restructuring of existing project portfolios and new initiatives. I INTRODUCTION i. Madagascar has just completed a process of political transition, with a democratically elected President and Parliament, and the formation of a new Government. The new Government is facing a much deteriorated economic situation and will have to take measures to put the country back on a growth path, which has been sadly lacking for the past twenty years. In order to help the country define its new strategies and policies, the World Bank produced a set of four documents, namely: an Economic Strategy Note 11 (ESN - November 1993), a Private Sector Assessment 2/ (PSA - December 1993), a Poverty Assessment (due in March/April 1994) and the present Agricultural Strategy Note. ii. The Economic Strategy Note makes the case for an accelerated growth scenario, and identifies agriculture as one of the key sectors that would have an essential role to play If GDP growth of six per cent per year is to be achieved. The agricultural sector not only dominates the economy, but also, as argued in this report, offers good potential for accelerated growth and the capacity to solve some of the country's most pressing problems-acute poverty, food insecurity in rural areas, and mounting environmental degradation. The objective of this report is thus to help the new Government define its strategy and agenda of actions for agricultural growth. iii. The report consists of three parts. Chapter One assesses past agricultural developments. Chapter Two analyzes their relationship with economic policies and identifies the sector's potential for growth. It also describes the major constraints that have prevented Madagascar from capturing that potential. Chapter Three focuses on a strategic agenda for growth. It argues that achieving accelerated growth will depend critically on a market-oriented agricultural strategy and on agricultural diversification and intensification. It reviews the three areas that must be addressed by the Government to return the country to environmentally sustainable growth with equity: restoring a non distortive price incentive structure; providing a rural infrastructure network conducive to trade and agricultural intensification; and rebuilding the capacity of the public and private sectors to offer responsive support services. iv. The report builds upon information and analytical work undertaken by the Bank in collaboration with the Government, and by numerous donors. The report integrates the conclusions of several sub- sector reports that have been produced in recent years: the Irrigation Sub-Sector Review (February 1993), an Analysis of the Vanilla Sector (December 1991), the Rural Financial Sector Review (October 1991), and the Rural Road Strategy Paper (May 1991). Additional studies, financed by the United Ntions Development Program (UNDP) and executed by the Government, have also contributed to this report: studies on Regional Specialization and Agricultural Growth (1990), on Agricultural Growth Linkages (1990), and on Agricultural Enusehold Production in the Highlands (1991). 1/ Madagascar, Note do Strat6gie Economique, Banque mondiale, Juin 1993 Z/ Madagascar - New Horizons - Building a Strategy for Private-sector, Export-led Growth - A Private Sector Assessment, draft report, World Bank, November 15, 1993 2 CHAPTER 1 PAST AGRICULTURAL DEVELOPMENTS I. ROLE AND PJACE OF THE AGRICULTURAL SECTOR IN THE ECONOMY 1.1 General. Madagascar, an island more than 1,500 km long, with a land areu of 587,000 kn2 (slightly larger than Franc,), is characterized by great diversity of climate, soils and topography. Madagascar's biodiversity is unique. The population is estimated at 12.4 million (1992), very unevenly distributed with an average 21/km2 but exceeding 100/km2 in the Central Highlands, and growing at a rate of 3 percent per annum. Close to one-third of the urban population of 3 million lives in the capital. With a per capita GNP of US$210 in 1991, Madagascar is one of the poorest countries in the world, despite its strong and varied natural resource base. Per capita Income declined by 40 percent in the last 20 years. The economy exhibited modest growth in the decade after independence lin 1960, stagnation during the period of inappropriate economic policies (1972- 82), resumption of growth under adjustment programs from the mid-1980s onwards, a downturn in 1991-92 with the onset of major political turmoil and strikes as of mid-1991, and a resumption of modest growth thereafter. Inappropriate economic policies (reflected in monopolies and monopsonies, price controls, shrinking and distorted public investments and services) were mainly responsible for the poor performance. With few exceptions, the annual rate of economic growth has been outpaced by the rate of population growth. 1.2 Dominant Role of Agriculture. Madagascar is a mainly rural society with about 80 percent of its population living outside cities, and agriculture providing the main source of livelihood. The dominance of agriculture (including agriculture, livestock, fisheries and forestry) in the economy is reflected in the country's main economic indicators: the sector contributes about 31 per cent of GDP (table 1 A); it supplies inputs to over half of the secondary (industrial) sector as well as to the tertiary (services) sector, employs more than 70 percent of the labor force; and accounts for about 60-65 per cent of export earnings (during the early 80s, agricultural exports represented 85 percent of total exports, and over 90 per cent if processed agricultural products were added). As table 1 B shows, the rate of growth in the secondary and tertiary sectors follows closely that of agriculture; hence, agriculture determines to a large extent the overall performance of the economy and, conversely, agriculture is directly affected by the economic policies pursued. 1.3 A Small-Farm, Subsistence Breakdown of Agricultural Per Cent Oriented, Agriculture. Small farm Value Added by commodity 1992 households (average farm size is 1.2 Rice 41 hectares) make for about 96 per cent Fishing 20 of the farming sector. They are Coffee, vanilla, cloves, pepper 13 largely oriented towards subsistence Fruits and vegetables 10 production: on average, only 20-30 Cotton, sugar, sisal 6 percent of production reaches the Other cereals, root crops 6 market. Table 2 shows agricultural Livestock 4 production in volume (table 2 A) and in value (Table 2 B). Of all Total 100 agricultural activities, rice growing is the most important, occupying about Source: Ministry of Agriculture, 1992 two-third of cultivated area and contributing about 41 percent of total value added of agricultural production in 1992. The fisheries sub-sector comes second in terms of value of production at 19 percent, with the small-scale traditional sector accounting for 75 percent 3 of total catch in v^lume terms. Despite a larger potential, the livestock sector is relatively minor, contributing ond) 4 percent. Agricultural processing dominates the industrial sector, as It produces about 70 per cent of the value of industrial output. The table below shows the relative Importance of the various sub-sectors. 1.4 Importance of Irrigation. About 1.1 million hectares are under some form of water control in Madagascar, the second highest in area irrigated among Afripan countries. This represents about 51 percent of all permanently cultivated land, and 80 percent of potentially Irrigable lands: the very high percentage shows that relatively little land can still be brought under irrigation. Some 72 per cent of all Malagasy farmers have Irrigated land, mostly in small plots averaging 0.84 hectare (Table 3 A). Irrigation ranges from simple water retention bunds on small fields to sophisticated networks in some medium and large schemes. About 26 percent of irrigated land is under schemes known as "classified" perimeters which have been developed and maintained by government, while the remaining 74 percent is under private schemes created and managed exclusively by farmers. Rice occupies over 80 percent of the land under irrigation, followed by sugar cane and cotton (Table 3 B). 1.5 Vanilla is Madagascar's second-largest source of foreign exchange-an average of US$48 million per year in the past few years, or more than 10 percent of total export earnings. Vanilla also provides the Malagasy government with substantial revenues, contributing a vital 10 percent to the national budget. Last, vanilla is one of the few sources of monetary income for large numbers of rural dwell.-rs in the northeast. Belonging, along with the Comoros and R6union, to a cartel known as Alliance de la Vanille, which imposed export quotas by the country of origin and import quotas by the purchaser country, Madagascar set high prices on vanilla exports and restricted the supply. This enabled Madagascar to extract substantial rents from the world vanilla market. 1.6 Regional Diversity. Ten distinct agto-ecological regions have been identified in Madagascar, defined by their climatic and natural resources characteristics. They range from humid tropical to low-rainfall desertic. Table 4 provides a brief assessment of the relative importance of agricultural production in each of the ten regions in 1989. The Highlands are the most important in nearly all food crops, although it remains the largest food deficit region and has only a limited share of the traditional export and industrial crops. The Lac Alaotra region, on the other hand, is the major surplus region for rice. By contrast, the eastern regions (Northeast, East, and Southeast) and the North are the main producers of export crops and tropical fruits; the North also dominates in the production of sugar. The westrn regions (Northwest and Midwest) predominate in the production of other industrial crops (peanuts, cotton, tobacco) and maize. Because of increasing dryness as one moves south and west, the Southwest and Southern regions focus mainly on crops with low water requirements (butttr beans and peanuts; maize, sweet potatoes and sisal, respectively). Among livestock activities, cattle and poultry provide the bulk of earned output (84 percent in 1989); cattle is particularly important in the South and Southwest where population densities are relatively low as well, and in the central highlands where it provides milk, manure and animal traction. Pigs account for another 14 percent. Within fishing, sea fish and shrimp production accounts for more than half of the value added (56 percent), and is mostly concentrated along the northern and north-western shores. Inland fishing maintains a significant share (a third) of the total but has been steadily declining in recent years. 1.7 Regional diversity is not only evident in climate and land use, but also in population density and ethnic composition. While average population density is low overall (20 inhabitants per sq.km), it ranges from 120 in the Highlands to 5 in the Southwest. Attempts were made in the past to "resettle" people from the highlands to sparsely populated areas. These efforts essentially failed because of the harsher climate, ethnic resistance and sociological factors (attachment to ancestral land) and have since been abandoned. 4 II. THE CHANGING POLICY ENVIRONMENT A. Policies Un.il the Early Eighties 1.8 Post-Independence Policies and Results. Madagascar enjoyed a period of sustained agricultural growth in the first ten years after independence (in 1960). Positive per-capita agricultural growth was fueled by appropriate macro-economic policies, favorable terms of trade, and the progressive development of agricultural markets and services. However, this growth process was highly dependent on external funding and expertise from the former colonial power, and did not improve Income distribution. Disappointed with the degree of political dependency and the slow pace of transformation in the economy, the authorities introduced sweeping changes and embarked on a new era of socialism in 1972. 1.9 The Socialist Period: Policies and Results. Despite rhetorical commitment to the public ownership of means of production, collectivization of smallholder agriculture was never implemented. Instead, the "socialist" policies focussed primarily on public control over agricultural marketing, finance and external trade, with the active discouragement of the private sector. Marketing for major food crops such as rice and edible oils, and traditional export crops (coffee, cloves, vanilla, and pepper) was nationalized; marketing boards with monopoly powers were established: stabilization funds reinforced; and prices set administratively. This system enabled the government to heavily tax the agricultural sector, in particular exports, either directly through explicit taxes or stabilization fund revenues, or indirectly (and increasingly towards the late 70s and early 80s) through appreciation of the real exchange rate. It has been estimated that, over the 1975-1983 period, for the three main exports of coffee, vanilla and cloves, producers received only 40 percent, 25 percent and 25 percent of the world market price, respectively. In the case of rice, domestic producer and consumer pices were controlled and kept below import parity levels, resulting in large public subsidies mostly to urban rice consumers reaching 19 billion FMG in 1981, equivalent to 2.5 percent of GDP. 1.10 Other important policy objectives included industrialization as the intended engine of growth, to which a large share of financial resources (both foreign and domestic) was devoted. This led to declining levels of public investment in agriculture and to reduced availability of consumer goods and agricultural inputs in rural areas. In addition, remaining investments in the agricultural sector shifted towards capital-intensive operations (state farms, irrigation schemes, processing facilities) with at best a mixed record of success, and into the rice sector, with little attention to the export crop sector. Finally, a local government refom carried out over the 1970s and early 1980s led to increased power for the central and city govei nments, and a decline in rural infrastructure investment and maintenance (roads, water supply). 1.11 As a result, agricultural growth declined to extremely low levels throughout the seventies (0.6 percent on average between 1970 and 1980), and turned negative between 1980 and 1983 (-0.1 percent). From 1974 until the mid-1980s, rice production stagnated as yields no longer improved and even declined in the early 80s. With continued population growth and higher urban per capita rice consumption resulting from cheap food policies, rice imports jumped from 20,000 tons in 1970 to 350,000 tons in 1982. Export crop production (coffee, cloves, vanilla and pepper) also stagnated in volume, and no new export crops were developed. B. The Reform Program (1984-1990) 1.12 Given the poor results of its past policies, Government initiated an adjustment program with support from the International Monetary Fund and the Bank, and introduced a set of adjustment measures between 1984 and 1990 aimed at removing major economic distortions. The measures covered exchanje rate devaluation, price and trade liberalization, and tax reform. Agriculture was 5 expected to play a major role in the adjustment program by providing the most likely source for the generation of export revenues and for food import substitution. Due to their gradual introduction, it was not until 1988 that the reforms began to create an environment conducive to the resumption of economic growth. However, following the political disturbances in 1990, this process was again interrupted: external trade liberalization collapsed in October 1991 with the closing of the Open General License (OGL) system of foreign exchange allocation, and the imposition of a 30 percent import surcharge in January 1992. 1.13 Agricultural sectoi reforms were gradually implemented, FIrst with rice (from 1983 to 1990), and then with agricultural exports (from 1985 until 1992). Trade liberalization and elimination of export taxes were implemented early in the reform process for non-traditional exports and pepper (19851988); for coffee and cloves, trade liberalization was initiated in 1988, combined with a tax reform 2/ in 1990, and eventually completed with the elimination of export taxes in 1992. Trade liberalization for the rice sub-sector was implemented in 1986, and the oilseeds sub-sector in 1988. In addition, starting in 1986, the Ministry of Agriculture (MinAgri) disengaged itself from the direct distribution of fertilizer ar ' other chemicals inputs. However, it continues to play a central role in the allocation of donated.' .'izer to established operators at prices below import parity levels. Also, important sub-sectors, such . vanilla, cotton, sugar, tobacco and wheat, have not been subjected to trade and price liberalization measures and remain controlled directly by State or public monopolies. Critical macro-economic adjustment measures complemented the sectoral reforms in 1987-1988 (55 percent devaluation in June 1987, establishment of the OGL in 1988, and extension of domestic price decontrol to all sectors of the economy in 1998), creating the conditions for a more open and liberal economy. Additional measures aimed at controlling public expenditures (public enterprise restructuring and public expenditure reviews) and reducing price distortions (tariff reform) were also initiated in i988. 1.14 These meiures gradually restored economic growth and led to a real increase in per capita income in 1989. However, the favorable economic climate established in 1988 was increasingly threatened by Government's inability to control buoget deficits.and by money creation A/ which, combined with an inflexible exchange rate management policy, led to a period of real exchange rate appreciation (between mid-1989 and mid-1990). The onset of the political turmoil in mid-1991 added to the economic slippage, resulting in the suspension of the OGL (October 1991), and the rationing of foreign exchange. C. The Impact of the Reforms on Agriculture and exports 1.15 Summary. As shown above, the reforms have been introduced in stages. They were not completed; their sequencing has at times led to contradictory signals to producers; and the renewed exchange rate appreciation over the past two years has partly reversed their beneficial effects. It is therefore understandable that the impact of the reforms has been mixed so far, varying across sub- sectors and regions. Nevertheless, the following trends are discernable: (i) fisheries and non- traditional exports, which benefitted from early trade liberalization and adjustment measures, 2/ The January 1990 reform significantly reduced the level of taxation for coffee, and replaced the multiple tax system for coffee and cloves with a single and progressive ad-valorem tax based on the declared FOB price for clove, and a reference price for coffee. In addition, for cloves, a minimum fixed tax of 1440 FMG/kg was set when the FOB price fell below US$2/kg. 4/ Money creation resulted from a large credit expansion in the two remaining Government-owned banks (including the agricultural bank, BTM), fueled to a significant extent by non-performing loans extended to public agricultural firms (at the end of 1990, SINPA alone accounted for about 10 percent of non-recovered BTM loans, and the six largest for about 20 percent of the total). 6 significantly expanded after 1983; non-traditional exports increased at an average annual rate of 20 madagscar - The Fisheries Boa percent during the 1986-1991 period, although 1970 9 1989. starting from a low base (see table 5); in the course of ....... to80 . the last decade fisheries exports became the second major foreign exchange erner; (ii) similarly, nn- I F 1 10,500 rice agricultural activities in the central highlands - sui scale fishing 6,400 8,500 52,000 where communications are more favorable (proximity s.,r".: u.*. do. prohaan .. . of large urban markets and adequacy of physical E*ux at Forit- infrastructure and agricultural services) - responded positively to improved price incentives (dairy products, potatoes, sweet potatoes, maize, barley, wheat); (iii) the traditional exports (coffee, vanilla, clove, pepper) and cotton sub-sectors, and their corresponding regions (East Coast, Northern region, West Coast) did not respond to the adjustment measures to a large extent because of a significant drop in world prices (coffee, cloves), the absence of full trade liberalization (vanilla), and the continued deterioration of communications; (iv) production of rice, after first taking off following the elimination of price and trade controls on rice trade in 1986, has also stagnated because of L poor market infrastructure (roads, markets and storage). Liberalization effectively corresponds to a reversal of cheap urban food policies, and has resulted in a considerably increased involvement of private traders and processors; higher farm-gate prices for paddy; lower per capita rice consumption with a switch to cheaper alternative foods; and the elimination of the rice deficit, with no imports in 1992, compared to imports of 150,000 tons in 1985; (v) the structure of exports, as a result of all of the above, has shifted dramatically away from traditional exports and toward increasing diversification. 1.16 Food Crops. The countrywide rice trade liberalization in 1986 encouraged many small rice traders and processors to enter the market and provide a more competitive trade environment, bringing two important benefits in the short term: an overall reduction of intermediary margins by about 25-35 percent and a reduction in seasonal price fluctuations by abu.Lt 20-25 percent. Moreover, the fears of Government that lifting price controls would lead to significant price rises in the "soudure" period proved unfounded: the security stock was operations only in 1986/87, and for smaller volumes than anticipated (18,000 tons whereas donors had pledged stock support of more than 100,000 tons). The reduction in intermediary margins especially benefitted producers through higher farm-gate prices (by 30 percent in real terms) whereas retail prices for rice in real terms stayed constant up to 1990. Most importantly, some supply response has been forthcoming: rice production in 1988/89 shot up by 12 percent (although less in subsequent years due to unfavorable weather and the persistence of marketing constraints), and rice imports have been steadily reduced, and were eliminated altogether in 1992 (although the reduction was mainly caused by a fall in per capita consumption and a shift to cheaper alternative foods such as maize and cassava). 1.17 While these achievements are real, they are also quite fragile. Throughout the reform period overall incomes, and especially urban inomes, have fallen and pressures on the Government to contain increases in food prices through some form of price and/or supply control are strong. The efficiency gains achieved through private domestic trading are probably durable, but domestic prices could again be depressed through cheap donor financed imports, especially at a time when foreign exchange is scarce and is being rationed. Yet, it is essential for farm-gate prices to remain attractive; this stimulus alone will incite (and make it affordable to) farmers to increase the use of cash inputs, so as to raise yields which are still quite low (1.8-2.0 tons of paddy per hectare, on average) and by these means achieve an increase in paddy production. 1.18 Traditional Export Crops. The adjustment measures taken to revive the export of coffee and cloves (elimination of the stabilization fund and export monopolies and return to private trading, 7 reduction and eventually elimination of all export crop taxation) did do away with the distortions that, in the 1970s and most of the 1980s, had caused production and export of these crops to stagnate. However, this failed to revive exports for cloves as its main market, Indonesia, has been irretrievably lost (domestic production now covers its needs), whereas for coffee, the world market price has fallen by 67 percent since 1987. As a result, although the reforms created some trade efficiency gains, current farm-gate prices for coffee and cloves are 10 percent and 20 percent lower respectively in real terms compared to 1982, and production continues to decline. 1.19 For vanilla, reform measures have been recently initiated, but the trade is still subject to controlled prices. MADAGASCAR VANILLA EXPORTS The consequences of past policies have EVOLUTION been dramatic: Indonesia has developed s0o as a strong competitor for Malagasy vanilla exports, and is offering much Ul sos lower prices for a vanilla that has not yet attained the best quality grades of 40 Madagascar but is rapidly improving. As a result, Madagascar's exports over 2 0 the past five years have been cut by two thirds and its share of the world market a on has fallen from 70 percent to 32 percent and is still falling. In addition, lower YEA export sales have led to a rise in unsold CMOK0 ow WET ----- FOS PIJCE (UMG) Government-financed vanilla stocks sJ M.M.a0r10"An. 4 v:M. 0MUYn- (reaching 4000 tons in early 1993, equivalent to eight times the average export volume over the past five years, and leading to the decision to destroy about two thirds of those stocks in 1993). 1.20 Non-Traditional Exports. Since 1986, there has been an important evolution in the composition of agricultural exports, in that non-traditional exports have steadily increased, both in absolute terms and as a share of the total. While in 1986 non-traditional exports as a group accounted for only 23 percent of the total, their share had grown to 51 percent in 1991, exceeding for the first time the value of traditional exports. Shrimp exports, which are the bigges, single commodity in this group, almost doubled in volume between 1984 and 1991, and now account for 30 percent of all non-traditional exports. However, a variety of non-traditional commodities, such as beans and butter beans, cane sugar, cashew nuts, cotton fiber, live cattle, hides and skins, maize, gherkins and litchis also developed a healthy positive export trend. As a group, non-traditional exports increased o, average by 20 percent per year in real value terms during the 1986-91 period. 1.21 Individually, the reasons for export growth in non-traditional commodities are quite diverse. Some are dominated by public sector parastatals that have increased exports due to depressed domestic demand (sugar, cotton fiber, cashew nuts); others have benefitted from foreign participation and know-how (shrimps, gherkins, litchis); and still others have profited from the opening up of regional markets (live cattle, hides and skins). Nevertheless, they have all benefitted from the 1987 exchange rate devaluation and from the liberalization of the trade regime, and demonstrated that a swift response from the agricultural sector to an open market policy is entirely feasible. 1.22 Conclusion: The Shifting Structure of Madagascar's Exports. With limited aggregate domestic demand (due to low per capita incomes), and a major import constraint, expansion and diversification of exports are critical for Madagascar's accelerated growth prospects. The more is 8 this the case because of the major shifts which have occurred in world markets since the mid-1980s, considerably reducing unit prices and volumes of Madagascar's three main traditional exports (coffee, vanilla and cloves). Over 1985-91, traditional exports declined in aggregate value from US$284 million to US$95 million, and from 65 percent to under 40 percent of total exports (see Appendix table 5). While in volume terms, Malagasy farmers' output expanded significantly in response to market and price liberalization measures in the late 1980s, the drop in the terms of trade adversely affected the aggregate value of traditional exports. This reduction was compensated by an expansion of non-traditional exports (notably shrimp and fish, fruit and vegetables, textiles). However, after achieving a small surplus in 1989, the trade balance turned strongly negative again in 1990-92. D. Impact of the Reforms on Rural Poverty and Food Security 1.23 Madagascar produces about 20 percent more food on a per capita basis than would be required to meet the FAO standard of 2100 calories per person per day. The main problem is therefore one of distribution; both in regional terms (between rural and urban areas and within rural areas), and in income terms. Within rural areas, food insecurity I varies from 54 percent in the dry southwestern province of Toliara to zero in the northern province of Antsiranana. However, in Madagascar, a higher percentage of urban dwellers are food insecure (42 percent) than of rural inhabitants (34 percent). This difference is even more striking when the comparison is made with reference to the minimum thresholds 0/: only 6 percent of the rural population falls below the minimum threshold in food availability, whereas in urban areas the proportion rises to 32 percent. The depth of urban poverty is explained by sharp income distribution differences; the top 10 percent of the urban population receives 80 percent of all urban incomes whereas the bottom 40 percent receives only 4.6 percent. 1.24 While information on the income distribution effect of the reform program is scant, the available evidence suggests that the urban poor have suffered most. Malnutrition rates in urban areas were significantly higher in 1992 than in 1984, a result of the elimination of controlled food prices, and the lack of overall growth in the economy. The impact on the rural population has been mixed; while farmers in rice surplus areas definitely benefitted, the income situation deteriorated for the majority of farmers who are in a rice deficit situation and have to buy additional rice during the year. However, domestic trade liberalization and tariff reform have benefitted the farming community as a whole, both by giving them an increased share of commodity sales prices and by allowing substitution into more profitable commodities such as dairy production in the highlands. For growers of traditional export crops, the reform measures have prevented an even further fall in rural incomes than they actually suffered but have nevertheless done little to improve their incomes. Thus, the gap in income levels between the Central Highlands and the East Coast area -where most of the traditional export crops are grown- has, if anything, increased. 5l Defined as the share of persons having access to 80 percent or less of the FAO standard of 2100 :alories per person per day. I Defined as the share of persons having access to 70 percent or less of the FAO caloric standard. 9 CHAPTER 2 AGRICULTURAL POTENTIAL AND CONSTRAINTS I. THE POTENTIAL FOR AGRICULTURAL GROWTH 2.1 Summary. A number of analyses carried out for this study all point to the conclusion that Madagascar has a solid potential for accelerated agricultural growth. In terms of production base, it has abundant arable land and a strong diversity among ecological zones, enabling it to produce a variety of agricultural products ranging from tropical to temperate; it has extensive irrigation infrastructure which if properly maintained allows it to produce crops at bigh yields and with sufficient flexibility to meet changing market demand; and given the present low yiela .vels for most crops, the possibilities for productivity improvement are huge. In terms of economics, it has a strong competitive base: present low labor costs allow it price advantages that enable it to substitute for agricultural imports, and successfully compete on international markets. In terms of ma availability, Madagascar has - with the exceptions of vanilla and cloves - only a tiny share in the international market for its products. This is also true for regional markets (e.g., South Africa, Saudi Arabia, Mauritius, La R6union) where the country's proximity and easier trade links would give it an advantage over its competition. In general, market potential remains far from being fully exploited. 2.2 Production Potential. Some 32.8 million hectares, or 57 percent of Madagascar's total land area, is estimated to be potentially cultivable. However, only 3.0 million hectares, or less than 10 percent, are annually cultivated, of which less than 2.0 million hectares is cultivated on a permanent basis. While the potential for expanded land use is large, three factors put strong limits on such an extension in practice: (i) much of the uncultivated land is ecologically fragile and the ongoing trend to expand cultivation in new areas is already having serious environmental consequences; (ii) ethnic considerations effectively prohibit a large-scale out-migration from the central highlands into lower population density areas, where most of the uncultivated land is found; and (iii) huge additional public investments would be required in rural infrastructure to access (and maintain access) to these areas at a time when sufficient monies are not available even to maintain the present road network. Conversely, already productive lands, especially in irrigated areas are capable of increasing production considerably. For these reasons, it would be more effective to focus public policies on incentives towards productivity improvement and crop diversification on presently productive lands, than on expansion of cultivated land. 2.3 The analysis on the state of agricultural technology (see table on yields and fertilizer use in Section II B) provides convincing evidence that there is room for large improvements in productivity. With the exception of rice, yield levels of most crops are presently below African averages. The application of improved husbandry techniques using only incremental labor, would already have a significant impact on productivity levels. In addition, cash input use of any kind is well below the African average (about half), providing broad opportunities to further increase yields. However, a large effort is needed to adjust internationally known technology to local conditions through adaptive research. Such a sustained effort could double yield levels for existing crops other than rice, and increase average paddy yields by about 50 percent. More and better technology is also required to reinforce the trend towards diversification into new crops, especially fruits, vegetables and livestock fodder. Physically and climatically, production of a broad range of new crops is entirely feasible. 2.4 Competitive base. Table 6 and Appendix A show the results of a comprehensive domestic resource cost (DRC) analysis carried out to measure the comparative advantage (or the degree of effective protection) of Malagasy agriculture vis-a-vis the outside world. The DRC analysis also 10 provides insights of comparative advantage of specific regions in Madagascar versus other regions. In summary, the analysis shows: a) a strong comparative advantage for nearly all of Madagascar's domestic food Commodity DRC Export crops, provided a competitive exchange rate is maintained; b) an equally strong comparative advantage Maize 1.44- 2.68 for a wide range of established export crops (robusta Rice 0.57 - 1.56 coffee, cocoa, vanilla, pepper, butter beans, and cashew Robusta coffee 0.29 - 0.82 nuts) as well as a comparative advantage in food and Arabica coffee 1.08 - 1.09 processed crops with export potential (peanuts, cotton, Cocoa 0.58 cassava, potatoes and rice); c) a comparative advantage Sugar, free market 1.29 - 3.34 for coastal areas in export crops and an advantage for Sugar, USA quota 0.46 - 0.76 internal regions such as the Highlands, the Northwest Cotton grain 0.19 0.65 and the Midwest to produce for home market needs; P 0.50 and d) an advantage for the Highland region to peanut oil 0.90 diversify into high value commodities for both home Cashews 0.98 and export markets, such as specialized fruits and Iproved potatoes 0.85-1.33 vegetubles, and dairy products; the outlying regions in Cassava 0.62 - 1.24 turn would supply the highlands with other agricultural Butter beans 0.40 - 0.58 commodities: ric-, pulses, maize and livestock feed from the Midwest; vegetable oils, cotton and sugar . from the coastal areas; and rice from specialized areas in the Northwest and Lac Alaotra. The DRC analysis different regions of the country) also shows the influence of high transportation costs on Source: Regional specialization and comparative advantage: better internal transport links agricultural growth in Madagascar, and lower port or airfreight costs would strongly AIRD, 1992 improve the comparative advantage of non-coastal areas for exports; in turn, inland regions are presently advantaged to serve the home market because of high unloading and internal transportation costs of imported produce. The level of comparative advantage would be even higher if the exchange rate was not over-valued. 2.5 Market Potential. Madagascar has traditionally been a low import-low export country with a high degree of self sufficiency. Agricultural exports and imports are 7 perctz and 10 percent respectively of total agricultural GDP. The value of agricultural exports in 1991 amounted to about US$ 290 million or US$ 24 per inhabitant, compared to an average of US$ 78 for Sub-Saharan Africa as a whole. With the exception of vanilla and cloves, Madagascar has had only very small shares of export markets for its products (e.g., 1.1 percent for coffee in the year of maximum exports, 1976). Even in regional markets, where its proximity would give it a cost advantage (e.g., South Africa, Mauritius, Saudi Arabia), its share has been insignificant. The world export market for high- value food products expanded rapidly, more than doubling over 1980-90 to total over US$100 billion by 1990. Three groups of food products accounted for the bulk of this growth: (a) fruit and vegetables expanded from US$20.2bn to US$46.7bn; (b) meat and meat products grew from US$14.8bn to US$27.7bn; and (c) fish expanded from US$4.8bn to US$13.9bn. This rapid growth has occurred in response to rising demand for higher quality (often fresh, off-season, or frozen) food products in increasingly affluent industrialized country markets. In the home market where most of the sale of agricultural commodities is concentrated, urban centers have grown but per-capita urban incomes, and hence per-capita consumption levels have declined. However, if this trend can be reversed, urban income growth will rapidly translate into consumption growth. Given its favorable cost/price structure, Malagasy agriculture is well placed to take advantage of market opportunities, both at home and abroad. 11 2.6 In the short to medium term, an export-led growth scenario for agriculture would offer the best chances for rapid growth given the large diversity of products that it can provide at favorable prices in an existing export market. As an Island, Madagascar is free from major plant and animal diseases; has a large diversity of medicinal plants and spices; and could capitalize on its unique flora and fauna to produce highly valued agricultural commodities for export. The strong growth and the growing diversity of non-traditional exports in the 1987-90 period as a result of liberalizing trade and the competitive exchange rate, give convincing evidence that such a scenario would be feasible. I. THE CONSTRAINTS TO AGRICULTURAL GROWTH 2.7 Summary. While Madagascar has a strong agricultural potential, it is also clear that it faces powerful and mutually reinforcing constraints to utilizing this potential. The major growth constraints can be summarized under four groups: a) inconsistent macroeconomic and sector policies harmful to agricultural development, and the resulting low level of effective domestic demand; b) lack of an adequate, market-friendly, legal and regulatory framework, resulting in pervasive intervention by the public sector, the absence of development of a competitive private sector, and the lack of empowerment of local rural communities; c) low technology use leading to declining soil fertility, deforestation and land degradation; and d) ineffective farmer support services and degraded rural roads and marketing infrastructure which do not allow markets to function efficiently, and an unbalanced public investment program that is of limited help in addressing the above limitations. Each of these groups of constraints will be addressed in detail below. A. Inconsistent Macroeconomic and Sector Policies. 2.8 The macroeconomic policies followed by the Government as well as the economic reforms carried out in the 1984-1992 period were already discussed in the first chapter. In summary, the main policy features were: (a) an exchange rate policy designed with a view to keep inflation down and satisfy urban consumers rather than to promote exports. This discriminated against agriculture and harmed exports; (b) high export taxation designed to secure fiscal revenue has importantly reinforced this trend; (c) price and trade policies were aimed at securing for domestic producers privileged outlets on the domestic market or providing cheap raw materials to domestic agro-processing industries; and (d) financial sector and credit policies still support oligopolies and inefficiencies, resulting in the limited availability of credit to the rural sector. Although the reforms undertaken eliminated many of the distortions, several important ones still exist, and are discussed below. The poor performance of agriculture, and of the economy in general, has resulted in a low level of effective domestic demand for agricultural products. 2.9 The renewed appreciation of the exchange rate in 1990-1992 has again been eroding the improvement in price incentives achieved with trade reform (1985-1987) and the devaluation of the currency (1987). The impact on aggregate performance has been swift, with real agricultural GDP growth slowing down to 0.5 per cent and 1.1 per cent in 1991 and 1992, respectively. 2.10 High export taxation existed till very recently in the case of vanilla, together with a highly regulated trade and storage structure. Here again the Government was not able to act due to conflicting fiscal and trade interests. Initial steps towards a complete liberalization have just been taken by the new Government, including the replacement of the previous single high export price by a floor price set at a lower level, the discontinuation of public financing of stocks, and the suppression of export quotas. It is too early to assess the impact of these measures. 12 2.11 Price and trade distortions still remain in areas where (publicly or privately owned) enterprises enjoy de-facto monopolies in processing (cotton, sugar, wheat, copra, oil palm, beef for export), and high levels of effective protection. They take a variety of forms: administrative pricing (sugar, seed cotton for domestic processing), high import tariffs (20 per cent for wheat, 45 per cent for copra oil, 65 per cent and 40 per cent for refined and crude palm oil respectively), non-tariff barriers (access to foreign exchange and credit for imports, regulatory restrictions to exports - such as for cattle on the hoof), or official monopolies (wheat, cotton). These distortions perpetuate inefficiencies and in some cases, such as dairy products and cotton, artificially low producer prices. 2.12 Financial sector and credit policies. The reform of the banking sector initiated in 1987/88 is not yet complete. The National Agricultural Bank (BTM), in particular, is facing major difficulties due to past deficiencies in cash management, and a lack of discipline and planning in general, and credit management which led to an uncontrolled expansion of credit and accumulation of arrears, and Government is now moving to privatize the institution. Although in general large farms and the larger marketing and agro-processing firms have good access to credit, the oligopolistic structure of both trade and the banking sector has allowed the banks to be quite conservative, resulting in difficulties for new firms to gain access to credit. In contrast to the good access to credit enjoyed by larger enterprises, the formal banking sector has hardly penetrated small-scale rural finance: only 1.5 per cent of small farmers in the country get credit, and smallholder credit represents only about 5 per cent of all agricultural credit. Cottage industries and small rural traders have virtually no access to formal credit. There is no extensive network of grass-roots level or informal financial institutions, leaving the majority of the population without access to financial services from banking institutions and therefore excluded from the resource mobilization and allocation system; this situation is now being addressed with the support of several donors, including the World Bank. 2.13 Low level of Effective Demand. Index of per capita income and Continuous declines in domestic per-capita privte cnsumtion1./income, exacerbated by an increasingly skewed distribution of income, have restricted demand 1972 1982 1992 for agricultural produce in the home market over the past twenty years. Data on national Population 7.1 9.2 12.4 income and consumption (table 7) demonstrate (million) the severity of the problem. Total GDP has GDP 100 73 6-1-hardly increased in constant terms between 1972 GDP 100 73 6and 1992; and total private consumption in that AG GDP 100 81 76 same period has gone down by about 9 percent, despite an increase in population from about 7 Cusumption 100 69 52 million (1972) to 12 million (1992). Thus, Source: National Accounts private consumption on a per capita basis, if indexed at 100 in 1972, stands only at 52 in 1/ in real terms 1992. 2.14 Data on past consumption trends for agricultural goods is scant. Consumption data on individual foodstuffs is incomplete and little is known about their weight in national household budgets. However, a "normal" pattern would be that the fall in per-capita consumption of agricultural produce would have been less thin the national average as people put priority on feeding themselves; also, demand trends for individual products would have differed depending on their price and income elasticities. Available data appears to bear out these assumptions: per-capita consumption of maize and cassava, for instance, which are low- income foods in Madagascar compared to rice, has somewhat increased over the past twenty years, by five percent and 10 percent respectively. For rice, however,; per-capita consumption declined by 30 percent from an all time high of 154 kilos in 1980 to an all time low of 108 kilos in 1992; for 13 sugar, it declined by 30 percent between 1975 and 1990; for dairy products, it declined by 15 percent between 1983 and 1989; and for cotton, demand from domestic textile producers declined by 20 percent between 1986 and 1989. From this admittedly narrow base, it would appear that overall agricultural consumption grew by less than 1 per cent per year whereas per capita consumption trends for individual products varied from an annual growth of 0.5 per cent to an annual decline of 1.5 per cent. 2.15 Stagnating effective demand has had important negative effects on farmer behavior. Farmers had little incentive to increase production for the market, and with stagnating or declining cash income they have not been receptive to the adoption of productivity raising techniques which would have implied the purchase of cash inputs. Farmers have thus concentrated on satisfying subsistence needs which, with an increasing population and little adoption of new technology, has meant considerable area expansion (in the order of 2.5 per cent per year) using traditional, generally low yielding, techniques. B. Lack of an adequate, market friendly legal and regulatory framework 2.16 Existing legislation is not adapted to encourage competitive development of the private sector in general, and in agricultural trade and production in particular. The PSA report highlights the problems related to the inadequacies of the Labor Code, the Commerce code, and the land tenure legislation, the lack of a Domestic Competition Law, the lack of transparency of the Investment Code and of the system of fiscal incentives, the need to overhaul/modernize the system of business law and the corresponding administrative and judicial system to encourage private investment, and the general lack of information readily available to the public concerning prevailing government laws and regulations. 2.17 Some of the important causes of the degradation of natural resources have to do with the lack of land tenure security, particuiarly in the case of tenants or share croppers, and the conflicts on the use of various natural resources (land, forest, water, fish) because of the ambiguous definition of resource use rights. There is a superimposition of traditional and modern laws and practices, and a clear need to revamp the current system. 2.18 The empowerment of local rural communities in commercial activities and in rural infrastructure maintenance is obstructed by the lack of suitable legal framework for farmer associations: the current provisions are based on the law on socialist cooperatives, and are no longer applicable. New legislation was recently enacted in the area of decentralized savings and loan associations; although this represents a major step forward, it is still too early to see an impact. New legislation has been proposed to provide a framework for producers to manage in common activities such as input supply, marketing of their produce, or water management and maintenance of irrigation perimeters (water users associations). The need for new legislation has also been identified in the area of professional and inter-professional organizations and in the area of quality management (norms, labels). 14 C. Low Technology Use, Deforestation, Declining Soil Fertility and Land Degradation 2.19 With land expansion continuing to use traditional techniques, and stagnating technology Average yields and fertilizer use adoption, one would expect the levels of kilogram per hectare - 1989/1991 technology used by farmers to be generally poor. The table in the box, which shows fertilizer use 002 Madagascar Aftican average and yields for some important crops as indicators of farm technology and compares them with on 0-1,0 African averages, confirms this picture with the seed cotton 400-600 1,000-1,200 exception of rice. The "better" picture for rice fritizer use masks the fact that indicated yield levels were already attained in the early 1970s and were at that time on a par with yields in Southeast Asia. In Asia, however, paddy yields have since reached levels in excess of 4 tons per hectare on average. 2.20 For rice, husbandry techniques (e.g seed conservation, nursery care, planting and weeding) are generally good to excellent in the central highlands, aided by long farmer experience and intense extension and research efforts in the past. Yield increases are however constrained by physical factors (lack of reliable water control) and economic factors (lack of cash input use). An example of what can happen if these constraints are solved is provided by the Lac Alaotra region: a rehabilitated irrigation network provides relatively good water control; most of the rice produced is for market; and the use of improved varieties and fertilizer is common. As a result, paddy yields average 3.6 tons per hectare or twice the national average. A similar example is found in small perimeters where rehabilitation has restlted in yield increases of 20-25 per cent. This is not to say that research is not necessary; tested an6 proven seed varieties yielding five to seven tons of paddy per hectare do not yet exist in Madagascar, and in this sense, a green revolution has not yet happened. Rather, the high yielding vari,ties that do exist are fragile and will produce their yield potential only with good water control and cash input use. As higher yielding varieties are gradually released by the research establishment (the International Rice Research Institute - IRRI - has a variety improvement program operating in Madagascar since 1982), the physical and economic bottlenecks to putting them to good use will become increasingly important. 2.21 For most other crops, however, the situation is fundamentally different. With the exception of wheat and barley (for which varietal and fertilizer research was funded by wheat and beer processing companies) and potatoes (for which a research program was carried out under a Norwegian funded project), the development of new technology has essentially been neglected, resulting in a sharp decline in the adequacy of existing techniques. Coffee varieties, for instance, have not been renewed and are gradually succumbing to disease; disease resistant cassava varieties are not available; hybrid maize is only now beginning to be grown on research stations; good quality fruit and vegetable seeds are not available either, and farmers have essentially only their own seeds to use. Several factors have been responsible for this neglect: the declining economic attractiveness of such crops due to lack of demand; Government marketing intervention or export taxation; the increasing isolation of the local research community from international research achievements due to foreign exchange scarcity; and increasing concentration on rice to the detriment of other crops, mostly because of lack of public funding. 2.22 The situation is similar in livestock production. With the exception of dairy in parts of the central Highlands and a small modern pigs and poultry activity around main cities, traditional no- input systems prevail. The situation actually deteriorated within the last 25-30 years due to the introduction of a number of economically important diseases that did not exist previously (e.g blackleg and liverfluke in ruminants, swine fever). This is due to the incapacity of public veterinary 15 services to control such diseases, mostly because of lack of funding and the existence up till recently of public monopolies on drug Imports and vaccine production. 2.23 A major consequence of the expansion In land cultivation has been an alarming increase In the degradation of natural resources, which has essentially taken three forms. In the central highlands, hillside cultivation using extensive methods has led to soil erosion (with loss of topsoil up to more than 100 tons per year on steep slopes), and extensive damage to roads, reservoirs and irrigation structures through siltation. In the drier northwest to southwest regions, the practice of bush-fires has expanded beyond the regenerative capacity of the plant cover, leading to serious soil cover reductions and fertility losses. Finally, forest destruction is increasing; deciduous natural forests along the West coast are disappearing at the rate of 2.5 per cent per year (80 per cent over the last 30 years) and humid tropical forests in the Eastern region at the rate of 1.5 per cent per year (50 per cent in the last 30 years), overwhelmingly because of the clearing of new land for cultivation. It is the forest destruction that has drawn attention of the international community to Madagascar's environmental plight; levels of endemism (fauna and flora groups that are found nowhere else in the world) are unprecedented, and the rate of forest destruction now occurring has caused the country to be classified as one of the ten "most threatened hotspots" on earth, and a major conservation priority. 2.24 As already mentioned, the causes for environmental degradation are several. The single most important cause of forest destruction has been the clearing of land for cultivation, and increasing bush-fires mostly find their origin in land use disputes between pastoralists and cultivators that go unattended. Commercial logging or mining has not been an important factor, and forest clearing for the gathering of firewood had only a limited and very localized impact (around towns). Forest clearing has always been occurring in Madagascar as the practice of shifting cultivation outside irrigated areas called regularly for the use of new land. However, higher population growth combined with economic stagnation over the last twenty years has accelerated the practice, destroying the balance between new land use and the regenerative capacity of abandoned land. Also, reduced effectiveness of Government services in the field has contributed to the destructive process; supervision of the use of gazetted forests has declined; rangeland management (maintenance of cattle trails and wells, grazing controls, limitation of bush fires) has essentially disappeared; and little has been done to clarify land tenure rights, leading to an increase in land disputes as well as in squatting and invasion of forest land. 2.25 The seriousness of the situation has brought about a heightened awareness of the Government and a massive response from the donor community. This has resulted in an Environmental Program under which, in its first phase, four lines of action have been launched: establishment and protection of 50 national parks and reserves; introduction of sustainable production systems in environmentally sensitive areas; accelerated titling of agricultural land (from 10,000 hectares to 50,000 hectares annually); and improved management of gazetted forests (covering an area about 50 times larger than designated parks and reserves). This program was started about three years ago, but achieving the goals set will take considerable time, as the main bottleneck in achieving them is the time needed to build implementation capacity on the ground. The program is initially concentrating on the protection of biodiversity, but in the longer term more attention would have to be given to other natural resources management problems, namely soil erosion, deforestation caused by shifting cultivation, and bush fires. 16 D. Inefficient Farmer Support Services, Degraded Rural Roads and Marketing Infrastructure, and an Unbalanced Public Expenditures Program (PEP). 2.26 Farmer Support Services are the core programs of the two ministries dealing with agriculture (the Ministry of Agriculture and Rural development, covering agriculture, livestock, fisheries and forestry, and the Ministry of Scientific Research, covering agricultural research). They include agricultural extension (under the agricultural production department), veterinary services (under the livestock department), irrigation (under the irrigation department), and agricultural research (under FOFIFA, a semi-autonomous institute). Agricultural input supply, previously under the Ministry of Rural Development, has been left to the private sector, with the ministry however retaining a role in allocating donor-funded fertilizer to private distributors. By and large, each of these services has obtained positive results in certain areas, such as a combination of good research and effective extension on rice in Lac Alaotra, and on wheat, potatoes, barley and dairy in the central highlands. However, they have generally not made a large impact on farm production and income. Their main problems have been: (a) a dispersal of actions under the influence of donor driven projects: five semi-autonomous extension services are operating next to the national extension service, in specific areas (e.g. southwest, middle west, southern highlands) or on specific crops (e.g. maize, cotton). Their internal organization varies (different salary scales and/or field allowances, different methodologies, different degrees of autonomy), their overheads are comparatively large, and coordination is difficult. (b) over-centralization: about 70 per cent of research staff operates out of the capital, resulting in little coordination between research and extension at field level, low execution of research programs, and inadequate feedback from farmers. In extension, the low field presence of qualified staff is apparent in the lack of quality control and lack of training of field extension workers; over-centralized financial management is also to blame, as all payment authorizations have to come from the capital and require substantial time and effort to get approved. Until 1992, all 40 senior Malagasy irrigation engineers were based in the capital; as a result, inadequate attention was given to maintenance of irrigation works, leading to a shorter lifespan of irrigation structures and, hence, to a low return on irrigation investments; (c) lack of current cost funding, perhaps the most important bottleneck to improved performance of ministerial farmer support services. The Recurrent Budget shows the overwhelming importance of the payroll (salaries) in total current expenditures (about 70 per cent of total). Increasingly tight budgets have led to longer delays in field allowance payments, and to lower allotments, giving field services little support to improve operations. 2.27 Rural Roads and Marketing Infrastructure. Perhaps the most important bottleneck for farmers to become part of the market economy is their difficult access to markets. Rural roads are degraded to the point that whole areas are cut off during most of the year, and local markets have ceased to function: out of a total network of approximately 40,000 kilometers, no more than 10 percent are estimated to be in acceptable condition. The existing program to deal with the problem is far from adequate: only some 900 kilometers per year were rehabilitated during the period 1990- 1993, while recurrent raintenance (mainly grading) is being performed only on approximately 500 kilometers of rural roads every year, and routine maintenance (pothole patching, clearing of ditches and culverts, grass cutting) on no more than 2,500 kilometers per year. 2.28 The central problem with resource mobilization for rural roads at present is the lack of funds fcr maintenance. Most decisions affecting rural road funding are being made by the central 17 government, even though the responsibility for rural roads has been assigned to the local governments. Although rural road rehabilitation will continue to require central funding contributed mostly by external sources, the effective decentralization of decision power and financial management is required to boost the mobilization of local resources and to increase the sense of responsibility for the maintenance of past investments. 2.29 Although rural roads have never been of optimal quality in Madagascar, a "cantonnage" or road-gang system existed in colonial and immediate post independence times, operated - and mostly paid for - by local or provincial authorities, which often charged user fees. These rural roads maintenance systems have essentially disappeared over the last twenty years as fiscal revenue collection as well as public works planning and execution have been increasingly centralized in the capital and concentrated on the main road network. Thus, the capacity of local administrations to keep up rural infrastructure as well as finance such upkeep has been run down to the point of disappearance. 2.30 With the decline in rural accessibility, the importance of rural markets has also diminished. Primary marketing has shrunk as traders increasingly concentrated on buying produce from farms with relatively good access, selling it directly on larger markets. As rural market trade contracted, so did market user fees and revenue, making it difficult for local authorities to maintain the market infrastructure or, even more, to invest in its modernization (e.g. refrigeration, storage). Some indication of this trend is found in the diminished number or frequency of rural markets over the past two decades and the increasing Importance of regional markets in larger towns. The main negative effects of these trends have been a lower percentage of farmers with market access, increased marketing margins and increasing isolation even of farmers with road access, owing to lack of market and price information. 2.31 Since the implementation of liberalization policies, however, most markets have steadily grown more dynamic, the degree of diversification and availability of products has increased, and competition is developing among collectors. Adequate marketing infrastructure is required to induce wholesalers, collectors and other important economic operators (such as future farmers organizations) to establish their business in market towns. The development over time of a network of specialized assembly, transit and terminal markets would improve the bargaining position of producers and increase the responsiveness of the sector to improved price incentives and demand signals. Studies carried out in South Asia have shown increases in the use of cash inputs by 25 per cent, the percentage of cash crops marketed by 35 per cent, and overall rural income by 20 per cent within three years after village access to markets was improved. 2.32 An Unbalanced Public Investment Program. The analysis of the Public Investment Program (PIP) for the agricultural sector over the past few years shows the traditional dominance of irrigation investments with more than 50 per cent of total; the importance of public investments in large scale agricultural plantations, such as for copra, palm oil, cashew nuts and sugar with 29 per cent of total; and of industrial crops such as wheat and barley, with 11 per cent of total. It also reflects the low priority given to such areas as food security or institutional strengthening. In recent years, however, an effort has been made to better coordinate or combine projects in sub-sector programs: for example, in 1985, the Government agreed with donors to concentrate its efforts in irrigation rehabilitation on small and medium scale perimeters, and to concentrate each donor's intervention on specific areas; in 1989, a national research program was adopted leading to better funding distribution and improved priority setting of research goals; and in 1991, a livestock sector program was agreed to on similar lines. The relative importance of irrigation is also being reduced, reflecting the end of investment on large perimeters; it represents only one third of the 1994-96 PIP, or one fourth if the environment sector is included in agriculture. 18 2.33 A major concern Is the 1994-96 Public Investment Program multiplicity and diversity of Agriculture and Environment projects for agriculture and (current billion FMG) environment (86 in the 1994- 96 PIP, as shown in box table Nbr of Amount % next page), and the projects corresponding dispersal of action. This is a reflection of Irrigation 9 135.8 24 (1) a general acceptance by Rural Development 15 70.6 13 the Government of all donor Agr. Research 3 27.7 6 project proposals; (2) the Livestock 6 65.6 10 absence of a coherent Agric. Services 16 71.0 13 agricultural strategy and Fisheries 3 6.1 1 policy framework that would Water and Forests 11 43.0 8 enable the setting of Environment 18 117.0 21 Investment priorities; (3) Miscellaneous 6 35.9 6 inadequate knowledge on Total 86 661.6 100 implementation results, which would be another factor in 5uc:Revue du PnFrizm gas=, ~ ~Investment selection; and (4) Secteurs do 'Agricultare ot de Environnement, ADE, December 1993 insufficient donor coordination. The pressure that has been brought to bear on Government to contain the overall PIP envelope in the face of mounting fiscal deficits, has merely resulted in cutting counterpart funds for individual projects (usually the larger ones) rather than in cutting out projects; the consequence is a growing number of under-funded projects and increasing delays in disbursement of donor funds. 2.34 Other important flaws in the PIP process are the following. First, actual expenditures cannot be compared with budgeted expenditures. Information on actual expenditures is presented in the aggregate and is late in being made available. It is understood, however, that (1) actual expenditures are significantly lower than budgeted, in the order of 30 - 35 per cent; (2) major delays occur in the timing of payments; and (3) non-payroll operating costs show the biggest shortfall, both in terms of timing and of amounts paid. Second, there are no systematic reviews by Government on the effectiveness of investments being undertaken. The work done in this field has been essentially donor-initiated, and of an ad-hoc nature (achievements of specific projects or, occasionally, of specific sub-sectors). They suggest that returns on sector investments have been generally low, with long delays in completion, defective operation after completion of the investment, and inadequate follow-up. They also still reflect the interventionist nature of Government policies in the past: a top- down approach rather than promoting initiative from the bottom up; a durable commitment to public involvement rather than allowing the private sector to flourish; and a far reaching centralization of human resources and decision making in the capital rather than putting most of them in the field. Irrigation investments, for instance, have shown a short life after completion and a low financial and economic return because of inadequate maintenance 2/; until recently, Government assumed the operational and financial responsibility for maintenance in all classified parameters, but was either not in a position to deliver, or delivered such services at high and eventually unsustainable cost. Public ownership of plantations has also had mixed to poor results, generally showing the inadequacies of Government run enterprises: high investment and operating costs, low production levels, hard to find markets, and the continued need for operating subsidies. Third, the lack if 2/ Madagascar - Irrigation Sector Review. World Bank, March 1993. 19 recurrent costs funding has reinforced these inefficiencies and contributed to the mixed performance even in the implementation of the Governments' stated policies. 20 CHAPTER 3 A STRATEGY FOR AGRICULTURAL GROWTH I. THE VISION 3.1 As explained in the Economic Strategy Note, future recovery of the economy of Madagascar will depend largely on the policies pursued by Government. A projection of current trends, which is the most likely scenario in the absence of a major reform program, would result in a continued decline of per-capita income. A return to a program of partial reforms, similar to the situation during the 1988-1990 period, would translate into very modest growth of per-capita income, such as it would take two generations to recover the 1971 level. Only a strong program of reforms would bring about accelerated growth that would be sufficient to recover the 1971 level in about ten years, and to double per-capita income in one generation. The high growth scenario presented in the ESN foresees growth accelerating to 6 percent p.a. by the year 2000. It is based on the development of diversified export activities with a high labor content, induced by an appropriately supportive policy environment. The 6 percent rate could be reached by combining a 4 percent growth rate in agriculture with a 10 percent rate in industry and a 6 percent rate in services. 3.2 The structure of the Malagasy economy is such that a rural development policy based on the intensification of the small-farm sector in the high-potential, densely populated areas of Madagascar will accelerate the development of the entire rural non-farm economy, increase the size of the domestic market, as well as provide sizeable off-farm employment opportunities. A more aggressive export strategy represents a powerful instrument for boosting agricultural production and incomes, but needs to be complemented with a strategy designed to satisfy the increased local demand for food resulting from increased incomes triggered by export expansion. A balanced public investment program would promote regional specialization and inter-regional trade, and hence achieve a more efficient utilization of the country's resources. Increasing productivity and diversifying production on existing lands would do much to slow down the expansion of land use under low productivity techniques which is now occurring and which is the primary cause for environmental degradation. 3.3 Madagascar can achieve agricultural growth of over 4 percent per year on a sustained basis; it has already done so in several years of its post-independence history. Agriculture has a strong comparative advantage both for import substituting crops and for export crops, provided t:e bottlenecks in channelling produce from farm to market can be solved, access to improved technology can be improved, and a competitive exchange rate which existed from 1987 to 1990 can be restored and maintained. It enjoys great climatic diversity and has more than half its cultivated land under irrigation. At current input and yield levels, the potential for increased productivity is high. 3.4 If tmis growth level could be maintained for fifteen to twenty years, the number of people living below the poverty line in rural areas could be reduced by two-thirds, food insecurity would decrease drastically, and environmental degradation through forest destruction much diminished as the pressure to clear new land for food production would become less acute. 3.5 Positive developments have already started in response to adjustment measures takea, showing that an accelerated growth path is well within the country's reach. In the 1986-1992 period, improved price signals to rice producers have led to increased deliveries to market and helped eliminate a rice deficit, that as late as 1985 had still resulted in rice imports of 150,000 tons, while achieving significant efficiency gains in the process in trade and processing. In the same period, farmers in the highlands began to diversify production in response to market demand: growing of 21 non-rice foodcrops (wheat, potatoes) and vegetables in the dry season on irrigated lands quadrupled in value, and dairy production tripled. Also, in the 1986-1991 period, the production of non- traditional exports grew by about 20 percent yearly, increasing their share in total agricultural exports from 15-20 percent in the early 1980s to 51 percent in 1991. With the right policies, these changes can be reinforced, broadened and expanded over much larger areas. 3.6 To address the constraints that have impeded overall development in general and agriculture growth in particular, the Government would need to move on a broad front. First, a package of macroeconomic policies would have to be adopted, aiming at stability (with inflation, fiscal and balance-of-payment viability) and the restoration of growth, most importantly through an open foreign trade regime and a competitive exchange rate. Second, a market friendly policy needs to be further pursued at the sector level, in which trade and prices are liberalized in the sub-sectors still under Government control (vanilla, sugar, cotton, wheat), markets are essentially left to the private sector, and foreign know-how and joint ventures with Malagasy firms are encouraged. A strong improvement in the efficiency of farm support services is needed, including a comprehensive decentralization to bring them closer to the field, both in management and decision making. For development to be sustainable, high priority is to be given to the proper management of natural resources. A large scale effort is required to improve rural infrastructure as well as farm to market links: better maintenance of irrigation structures and farm roads; rehabilitation of the rural and provincial road network; and better telecommunications. To effect these changes, a full revision of the public investment program will be required. The measures to be taken to implement these policies are detailed below. II. THE POLICIES FOR GROWTH A. Market-Friendly Policies 1. Economy-wide Policies 3.7 In order for the country to reach and sustain higher growth levels, an appropriate package of racroeconomic policies would have to be adopted, as discussed in the ESN. It is crucial for alagasy policy makers to realize that agriculture policy is an integral part of economic policy and to take more account of the impact of macroeconomic policies on agriculture, especially trade, taxation, and exchange rate policies. There is ample evidence that this was not the case in the past. The tenuous links between macro-economic and agriculture sector policy formulation are not surprising. The uroan bias has traditionally been strong, fiscal concerns have been dominant, and the representation of agricultural concerns in the government has been traditionally weak. Rice trade reforms were agreed upon, but only after assurances were given that urban consumers would be spared strong price increases through the setup of an intervention stock mechanism. Coffee and cloves export taxes were reduced and finally eliminated only after precipitous declines in world market prices had occurred, and vanilla export taxation has had a highly distortive effect on exports and farmer income. The renewed appreciation of the currency from 1990 onwards happened at the same time as some of Madagascar's main agricultural exports (coffee and cloves) experienced their worst decline in world market price levels, thus dealing a double blow to their profitability. 22 3.8 The following are the main policies which the Government have to pursue in order to create and maintain an economic environment which would provide proper incentives and the conditions conducive to private sector development: (a) an open exchanga and trade regime guaranteeing the continuing competitiveness of the country's exports and free access to foreign exchange for goods and services as well as capital operations; (b) a tariff and tax policy which does not discriminate against agriculture; limits export taxes to vanilla; and in general restructures the customs tariff to eliminate distortions among sectors; (c) fiscal and financial policies (reduction of the budget deficit; improvement of the banking system) which make credit more readily available to the private sector, thereby facilitating private investment and export diversification; and (d) a framework of domestic competition and regulatory policies aiming at improving the business environment by removing inequalities of market access, especialy for Malagasy small entrepreneurs and for foreign investors, and reducing risks and transaction costs in doing business in Madagascar. 2. Sector Level Policies 3.9 In order to create conditions conducive to private sector development, the Government would have to limit the State's role in agriculture to orientation, guidance, and the provision of a proper legal and regulatory environment, and would promote fair competition by establishing appropriate trade policies as well as by withdrawing from public ownership or majority partnership of production, processing or marketing enterprises. 3.10 Key aspects of the legal and regulatory framework that need to be improved include (i) a framework for producers to manage in common activities such as input supply, marketing of their produce, or water management and maintenance of irrigation perimeters (water users associations); (ii) professional and inter-professional organizations; (iii) quality management (norms, labels), particularly in relation to exports; (iv) land tenure laws, in order to clarify natural resources use rights, provide stronger tenure security, facilitate the development of land markets, and allow land possession to all those willing to use it productively and sustainably including, under certain conditions, non-nationals; (v) a framework for public and private entities to jointly manage activities of common interest such as public infrastructure; and (vi) the elimination of unnecessary export restrictions, limiting them to cases of protection of cultural heritage or of endangered natural resources. 3.11 Public monopolies exist in a number of areas benefiting from administrative pricing or trade restrictions, as well as budget transfers enabling enterprises to survive inefficiently. Examples are: (a) the cotton company HASYMA, which maintains a ginning monopoly as well as a two-tier pricing system (for local requirements and for export); (b) the wheat company Kobama which maintains a wheat milling monopoly; (c) the two sugar companies SIRANALA and SIRAMA; (d) the copra oil company SOAVANIO; and (e) the palm oil companies ANTALAHA and SOMAPALM. While some of these are not officially monopolies, all either enjoy a protected market behind trade barriers, and are allowed high administrative prices, together with barriers to entry of competitors, or have preferential access to credit. More market-based policies encouraging competition should be pursued in those markets, eliminating trade barriers and barriers to entry, as well as budget 23 transfers to such companies. Price stabilization funds should be limited to exceptional circumstances, and with full financing by the industry; preference should be given to price stabilization through access to futures markets whenever feasible. Finally, these market-friendly policies should be complemented by partial or total privatization of the respective state-owned companies. 3.12 In the case of vanilla, initial steps towards liberalization of the trade, storage and pricing regulation framework were taken recently, but need to be followed by further action. The policy is to move progressively towards free pricing and a fully liberalized free trade, on the basis of a single ad valorem export tax (with mechanisms for its adjustment); the establishment of an auction system at export; and the constitution of an interprofessional organization to take over from government quality control and market promotion, while future stocks would be privately managed. 3.13 In all cases except for vanilla, preparatory work is needed to devise the specific measures for action. This work would cover: a review of tariff and import restrictions; company audits to determine net worth and profitability; review of state support in recent times, including budget transfers; review of stabilization funds where applicable (sugar, cotton, wheat); and review of the constraints to access to futures markets. In the case of the cotton sub-sector, a review needs to be done on the impact and results of the reorganization carried out three years ago, in order to develop and implement action plans to transfer Hasyma's seeds production and extension activities (i) to organized private sector in the North-West; and (ii) to the regular Ministry of Agriculture extension service in the South. In the case of sugar, a study is required to establish an appropriate pricing policy to cope with highly distorted and unstable world market, establish price-setting mechanism for cane at farm-gate, and liberalize trade. 3.14 The focus on trade liberalization and privatization of public companies has led to concerns that, in view of the preponderance of underdeveloped trade structures and markets, public monopolies would be merely exchanged for private monopolies, and private traders would benefit excessively. However, such monopolies would be exposed to competition by liberalizing imports and facilitating new entries, and by expanding market knowledge to all participants. B. Improving Farm Support Services 3.15 To pursue sustainable agricultural growth, farm support services need to be given a new mandate, as well as reorganized and strengthened. The new mandate embodies the recognition that farmer support services have a key role to play in helping to protect the environment and indeed the sustainability of agriculture itself through intensification and productivity growth. The strategy would promote a commodity-neutral approach rather than providing support to the development of specific crops or commodities. Such an approach would require from the agricultural public institutions a different set of skills and fundamental changes in their role and functions. It would involve concentrating mostly towards advisory and service functions, and a more decentralized institutional organization. A reorganization is necessary to decentralize manpower, decision making and financial management to bring them closer to the field; to review salaries and allowances to make it more attractive to qualified people to work outside the capital; to establish a comprehensive training program to improve qualifications and experience; and to promote better linkages both among services and between services and farmers. 3.16 Central functions would remain limited to sector-wide policy issues (pricing policy, size, sectoral and geographical composition of the public investment and expenditure program, aid coordination), and operational functions would be effectively decentralized and coordinated at the local level, combined with a decentralization of the planning and programming functions. At the 24 central level, this requires strengthening of the public sector capacity to monitor, evaluate and analyze developments in agriculture, and to prepare coherent agricultural public investment programs. 3.17 At the decentralized level, regional support centers should be strengthened through Improved autonomy in decision-making, budget control and accountability with a view to coordinate and adapt actions at the local level; their role would be concentrated on advisory and technical services to the private sector, including farmers organizations, NGOs and local governments. Planning and programming functions would be decentralized at the regional level; the provision of centrally- provided financial resources to manage and maintain local investments would be based on national sectoral and geographical priorities designed at the central level, and would be dependant on the capacity of local institutions (local governments, farmers associations) to contribute with locally- mobilized resources. 3.18 In research, a restructuring of the ongting multi-donor supported FOFIPA program is being completed, which will help to improve the quality of research. The restructuring aims at providing increased autonomy and control of incentives to FOFIFA to effect the changes that are needed but that have met with resistance from the traditional research establishment. The major elements of the restructuring are: (a) major emphasis on multidisciplinary field-based research teams; (b) research to be funded by individual contracts to increase its relevance to farm needs and the accountability of the researchers; (c) a participatory system to conceive and approve research proposals, which includes farmers, extension, industry and researchers; (d) a prioritization of the regions to be covered initially and a plan for gradual expansion. 3.19 Although the Government's objectives of decentralization coincide in principle with the above strategy, it has not yet been possible to coordinate the step-wise supply of basic infrastructure and social services to agreed priority areas, and conditions in the field remain very difficult. In contrast, donor coordination is beginning to show results. The majority of the donors involved with FOFIFA have agreed to support the management objectives and to provide financing on the basis of the priorities set by the institution in the participatory framework, instead of continue to create and maintain isolated centers which create divisiveness and discourage team work. 3.20 Improving the extension sector will be more complicated than research, as there is no single entity in charge, responsibilities are divided (sometimes by crop and sometimes by region), methodology and internal organization differ importantly, and pay scales (especially field allowances) vary. In addition, in extension, unlike in research and irrigation, donor coordination has not been effective, and many donors have been involved in promoting their own projects, rendering the restructuring task even more difficult. On the other hand, there are a number of NGOs and some agro-business companies that carry out extension programs. In the search for the establishment of a national professional agricultural extension service, the Government should continue attempts to rationalize the multiplicity of agricultural extension initiatives in the rural areas by concentrating its efforts on ensuring that a maximum number of the country's farmers have access at a minimum to one multidisciplinary source of advice. Whatever the source of funding for such extension workers, there should be a uniform effort to ensure that they benefit from regular training, have access to technical subject matter specialists on a regular basis and are equipped and sufficiently mobile to carry out their task efficiently and cost-effectively. They should be relieved of peripheral responsibilities such as input distribution, credit and statistics collection to enable them to concentrate fully on their principal task of technology transfer. 3.21 In view of the complexities and sensitivities involved, it is proposed that, first, an assessment of current experience be carried out by the Government which would for the different extension units (a) examine methodologies used and their effectiveness, in terms of farmer adoption rates and extension agents' back-up support and regular training by supervisory staff; (b) examine present 25 staffing in terms of coverage ratios (field extension agent to farmers, supervisory and scientific support staff in field to extension agents and between field and headquarters, administrative staff as proportion of total), and in terms of training and competency (experience, farmer appreciation); and (c) examine the different financial and administrative arrangements in place, determining which ones work well and analyzing the factors that make them work or not work. A possibility would be to carry out an inventory of the organizations conducting extension, and then convene a series of workshops, to be followed by a national conference, including the participation of farmers and research, and possibly donors representatives, to rationalize the principles and to draw on the best practices of all organizations. Diversity Is not objectionable; what is desired is to promote efficiency and effective use of resources. Based on its findings, the conference would make detailed recommendations on the respective roles of the various parties involved (public versus private, links with research, role of farmers organizations, etc) and outline proposals on changes needed to achieve a better streamlined and more effective extension service. All on-going and future publicly funded projects would then be expected to take.the recommendations into account, including, if need be, by restructuring or closing down some operations. 3.22 Finally, for the provision of agricultural inputs, distribution of fertilizer and agro-chemicals has been in principle privatized since 1986. However, large state-owned trading companies still have a large share of fertilizer distribution, and the distribution of donor-funded fertilizer is often carried out !n a non competitive arrangement which prevents the development of private sector channels. There is a need to establish transparent rules for the management of aid-in-kind. A formal mechanism should be defined to coordinate deliveries in order to prevent excessive concentration of such imports at any given time, and to make sure that the price of the products to distributors reflects their market value. The problem with fertilizer is not only availability, but also formulation. The formulation of fertilizer mixes available on the market should correspond to the needs of the soil. If not, farmers attempting to supply adequate amounts of the deficient elements are also forced to apply unnecessary ones, thus raising the cost of fertilizers. 3.23 Also, the Government maintains seed stations for the multiplication and sale of improved (largely rice) seeds which are poorly managed, have low sales volumes and need budget support to continue operation. A restructuring program is recommended whereby Government would cease to operate seed multiplication stations and would, instead, promote seed multiplication by private farmers, assuming only a quality control role (through seed certification). C. Improving Natural Resources Management 3.24 Deforestation, soil erosion and, more generally, the degradation of natural resources are complex phenomena that result from the interaction of a number of economic, social, political, historical and natural factors. A participatory approach is required to develop appropriate mechanisms for raising community and farmers' awareness of the natural resources degradation problems and determining priorities in optimizing the use of available land resources. NGOs should play a crucial role in helping rural communities design and implement sustainable management plans for their land and other natural resour-es. The Government should support such an approach by improving the knowledge of natural resources (forests, water, land, biodiversity) to provide the basic understanding upon which they would be managed; by implementation of progressive land use reform that would lead into more secure distribution of productive land and more efficient soil conservation and agroforestry practices and surface water use; and by providing training and financial incentives (e.g. sharing of revenue from logging). 3.25 Four specific problem areas need to be reviewed and on-going programs need to be strengthened: grazing management and the problem of bush fires; forest management; fisheries 26 resources management; and land tenure security. If properly strengthened, these programs would have a major impact in protecting the environment while at the same time providing better land possession and user rights to farmers, enabling them to invest in longer-term land improvement. Restructuring of Government services dealing with these sub-sectors has alreAdy been under way for some time; however, some crucial elements have been left aside, and need to be tackled. 3.26 Regarding grazing management, the recent reform of livestock services has basically left aside the question of land management in the dry western regions. The causes of the widespread occurrence of bush fires in these areas, which have been increasing in frequency and intensity over the past decade, are not well understood; as a result, it is not possible to set up an action program to deal with this environmentally destructive phenomenon. It is believed that one of the common causes is an increase in land disputes arising from new settlement in pastoral areas and the increased use of water sources for agriculture, restricting access of pastoralists to water, whose grazing rights are not always clearly established. A study is urgently needed to determine the extent of these developments and their impact on environmental degradation, and more generally to determine the various causes of the problem, so that counteracting measures can be devised and implemented. 3.27 In forest management, significant progress is being made in the establishment and protection of national parks and reserves; however, the management of gazetted forests which are about 50 times larger than the area under parks and reserves and which are being degraded by uncontrolled cutting and cultivation, is totally inadequate, nor is there a coherent view on what needs to be done and where the responsibility for it should rest. The policy should be based on the premise that (a) gazetted forest land in many cases overlaps with established user rights of rural dwellers, and that the separation of ownership cannot be solved by legal action alone, but needs local consensus; (b) better controls on cutting as well as higher user fees need to be established in forest areas without disputed land rights so as to discourage invasion and squatting; and (c) the forestry department alone will not be in a position to carry out these tasks; much of this forest management work needs to be contracted out to the private sector or NGOs. 3.28 Fisheries resources management. In the case of shrimps, an increase in the number of fishing vessels accompanied by a decrease in catch per vessel indicates that production has reached the maximum sustainable yield and that there is a risk of over-fishing. What is required, not only for shrimps but also for other fisheries (lobsters, crabs, demersal fish, etc) is establishing a proper management regime. The corresponding actions include: (i) better assessment of stocks, on a continuing basis; (ii) strengthening fishing monitoring by Government; and (iii) formulation of management plans (i.e. maximum catch and allocation of fishing rights) and a permanent discussion mechanism. 3.29 In land tenure security, work has started on increasing capacity of the land titling department of the Ministry of Agriculture. However, recent studies in Africa indicate that, depending on specific circumstances, formal land titling is not necessarily the best approach to land tenure security. A review of the various land tenure systems, laws and practices in the country should be undertaken, with the objective to define better tenure security procedures for owners of traditional rights so that they no longer constitute a brake on incentives to maintain soil fertility; and to provide more legal security for share cropping arrangements. An implication is likely to be the need to strengthen the judicial system in rural areas to better deal with land and water ownership disputes, and to strengthen field services to carry out boundary establishment and cadastral work. 27 D. Infrastructure 3.30 As described earlier, improved market access critically depends on rural roads and local markets adequate maintenance, which in turn hinges on the capacity of local authorities and adequate financing. As part of the country's move towards decentralization, a new definition of responsibilities and distribution of fiscal resources still needs to be worked out. To set up an investment program for rural roads rehabilitation and establishment and/or modernization of local markets will pose considerable problems. Local authorities are neither equipped for road maintenance nor can they currently raise and earmark funding for such purposes, as road planning and execution has been exceedingly centralized; labor intensive technology and the use of local contractors would be the most appropriate methods to implement such a program, yet the technology is only beginning to be reintroduced in Madagascar and the local contractor industry is weak and in considerable need of training and equipment finance; and current Government funding is not even sufficient to maintain the main road network, which has usually had priority over secondary roads. These difficulties will, however, have to be overcome. Based on experience with rural road programs elsewhere in Sub- Saharan Africa, successful approaches usually contain the following elements: (a) the setup of a specific rural roads agency, the statute of which allow it to have ministerial representation (from both Agriculture and Public Works) and NGOs active in rural road building or maintenance which can ensure broad support from central and local governments, private interests and the donor community, and help local communities create maintenance programs; and (b) the setup of a rural road fund which can act selectively throughout the country wherever local communities are prepared to organize and fund rural road maintenance after rehabilitation; and (c) close collaboration with agencies such as the International Labor Organization (ILO) which are familiar with labor-intensive methods. 3.31 In irrigation, the strategy for development of the sub-sector and the role of the state is changing. There is a growing awareness of the need for the disengagement of the state in the operations and maintenance of public irrigation schemes. The diminishing public resources available for the purpose and the past failures of large scale irrigation schemes run by parastatal companies have led the government to search for ways to delegate responsibility for operations and maintenance to farmers and water users' associations. There is also a growing awareness of the heavy costs of irrigation maintenance created by the environmental deterioration of catchment areas and poor drainage of existing schemes. Proposals exist for changing the role of the Rural Works Directorate from one of operator of public irrigation schemes to one of advisor to the water user operators of the schemes. There are also proposals to strengthen the environmental assessment capacity of the Directorate in response to the need to minimize the negative effects of environmental damage during the process of rehabilitation and operation of irrigation schemes. The need to decentralize the Rural Infrastructure Department in charge of irrigation, so as to enable them to focus better on their task of helping farmer organizations, is now fully recognized. There is also a growing awareness that the benefits from the improvements in water management can be multiplied if accompanied by an intensification of the associated activities of the agricultural extension and research and development services. Fundamental to the implementation of these new initiatives in the irrigation sector is the need to harmonize the funding proposals from the external donor community to ensure that the new irrigation strategy is being applied consistently. To this end close consultation between Government and outside donors needs to be maintained on a regular basis. E. Reshaping Public Expenditure 3.32 In order for the PIP to have a discernable impact on agricultural growth, piecemeal improvements will fall short of being effective. PIP reviews of sector expenditures in the past have generally concentrated on the soundness of individual projects, and poor investment decisions have 28 by and large been weeded out. What is needed now, is a fundamental reorientation of agricultural investment policies, on the basis of the proposed strategy, and a much greater focus on effective implementation. The reorientation should not be limited to the agricultural sector per se, but should aim at securing the geographical coordination of public investment In different sectors designed to alleviate generic constraints to agricultural growth (rural infrastructure, public services, private sector development), with priority to be given to the high potential regions of Madagascar. 3.33 In addition, recurrent expenditures would have to be increased and reoriented in line with the strategic objectives, so as to provide better farm support services, better maintenance of infrastructure and more effective protection of natural resources. Both investment and current expenditures will have to be accommodated within a prudent overall fiscal framework that would be designed to reduce the deficit over time, and thus foreign aid dependence. However, there may be scope, temporarily, for even higher cost sharing by the donor community than is currently the case, in order to facilitate the strategic reorientation of public expenditure. III. A PROGRAM FOR ACTION A. Preparation and Dialogue 3.34 Any program for action needs to be underwritten by both the Government and the stakeholders on a broad basis (policy makers, managers and implementers, farmers, processing and trading interests), for it to be ultimately successful. For this to happen, the concerned parties need to have an active role in devising it. The first step in this process is to arrive at a consensus on the overall strategy and policies as these will form the ultimate criteria for determining priorities and for allocating public expenditures. Moreover, criteria need to be specific so as to facilitate subsequent decisions and to avoid a multiplicity of undertakings in different directions. At present, a forum for discussing such policies does not exist and it needs to be created with some urgency. It would be highly desirable to establish an agricultural policy unit within the Agricultural Ministry, with a mandate to (a) advise the Minister of Agriculture on the impact of economy-wide policies; and (b) devise clear sector policies to achieve stated goals. As different ministries deal with or affect agriculture - including the economic ministries, an interministerial committee of high officials below the ministerial level should also be formed. This committee should not be limited to Government officials and agencies; it should include representatives of agricultural interests outside Government, such as trader and enterprise syndicates, farmers unions and NGOs. In time, such a committee could become an important support group for the Ministry, and vice versa, the Ministry could become a more effective supporter of such groups. 3.35 Finally, the Government needs to communicate its strategic thinking to donors and guide donor support along the newly established lines. The continuation of past strategies has led donors to continue to invest in projects that were known to have poor results; conversely, donors' preference for financing investments over operating costs has caused neglect of the capacity building that is necessary to make investments work. As 82 per cent of the PIP is donor financed, Government dependance on donor funding is high and the latter's full endorsement of a development strategy is crucial. B. The Proposed Approach 3.36 As a first step it is proposed to concentrate on the establishment of the policy group mentioned earlier. Once established, the group should review and discuss the strategy paper recently prepared by the Ministry of Agriculture, as well as the present note and any relevant study (several 29 donors have their own strategy/policy paper for Madagascar). Emphasis should be placed on the analysis of the agricultural sector; the impact of the reforms that have taken place; the constraints to growth and the proposed policies and instruments proposed. Once the group has reached a consensus on growth policies to be followed, the Public Expenditures Program for the sector should be subject to a comprehensive review in order to adjust It to the new priorities. At some point during this process, the Government should invite the donor community to support the new strategy through the restructuring of existing project portfolios and new initiatives. 3.37 If the process described above can be followed through, it would clear the way for a coherent policy framework on which strong local commitment would be acquired. It would open the first real opportunity to make structural changes in the Public Expenditure Program, that are now beyond the reach of Government, because of the project by project approach followed so far. If the policy debate is conclusive it would at the same time change for the better the internal procedures now being followed in the Ministries to arrive at the PIP, whereby each department submits its proposals in isolation without sufficient guidance or sense of priorities, leaving it to the ministerial level to bring them in line with the criteria set by the Economy and Planning Ministry. Together, these changes would do much to reestablish a sense of mandate and internal commitment that has been lacking in recent years and without which no development undertaking can be successful. 3.38 Whether the above approach will succeed, will depend on the coherence of the policies proposed; on the local recognition of the need for institutional change; and on the willingness of the donors to go along with a different way of funding development actions. The policies proposed would need to bring (a) an awareness of the crucial link between agricultural growth and favorable macroeconomic policies; (b) on the need to promote productivity development and reign in land expansion, as a means to limit environmental damage and, at the same time, attain agricultural growth in a sustainable way, i.e. without destroying the land base; and finally (c) a realization that improved techniques will be adopted by farmers only if they can get higher profits. For the first to happen, market-friendly policies need to be followed; for the second, a major revamping of farmer support services in research, extension, irrigation maintenance and input supply is needed; and for the third, a local capacity building program for rural infrastructure rehabilitation and maintenance will be required. Each of these will have major implications, especially in financial management and in shifts between investment and current cost funding, and would require active donor support to succeed. Also, most will need agreement and active support of private and regional interests outside Government as the latter will not be able to go it alone in terms of funding and human capacity. They will however render large benefits to the country, in an acceleration of growth and an increase in per capita income, in an alleviation of acute and widespread poverty, and in a reduction of food insecurity. TABLE 1 A. NATIONAL ACCOUNTS SUMMARY (Billions of FMG in 1984, constant prices) 1984 1985 1986 1987 1988 1989 1990 1991 1992 AGRICULTURAL SECTOR 531.0 536.9 554.3 568.2 580.7 610.9 623.6 628.8 633.8 SECONDARY SECTOR 195.0 197.6 205.1 214.8 218.6 221.3 219.1 216.8 212.9 TERTIARY SECTOR 808.1 815.1 820.7 819.3 855.6 890.7 925.8 854.3 861.9 GDP FACTOR COST 1504.6 1521.9 1551.9 1570.2 1623.6 1689.8 1734.7 1672.7 1687.0 Net Indirect Taxes 190.4 192.7 196.3 198.6 205.4 213.8 227.6 164.2 167.8 GDP MARKET PRICES 1695.0 1714.6 1748.2 1768.8 1829.0 1903.6 1962.3 1837.0 1854.8 Agriculture as % of GDP 31.3% 31.3% 31.7% 32.1% 31.7% 32.1% 31.8% 34.2% 34.2% o B. ANNUAL RATES OF SECTOR GROWTH 1984 1985 1986 1987 1988 1989 1990 1991 1992 AGRICULTURAL SECTOR 1.1 3.2 2.5 2.2 5.2 2.1 0.5 1.1 SECONDARY SECTOR 1.3 3.8 4.7 1.8 1.2 -1.0 -1.1 -1.8 TERTIARY SECTOR 0.9 0.7 -0.2 4.4 4.1 3.9 -7.7 0.9 GDP FACTOR COST 1.2 2.0 1.2 3.4 4.1 2.7 -3.6 0.9 Net Indirect Taxes 1.2 1.9 1.2 3.4 4.1 6.5 -27.9 2.2 GDP MARKET PRICES 1.2 2.0 1.2 3.4 4.1 3.1 -6.4 1.0 Source: World Bank Standard Tables TABLE 2 A. VOLUME OF AGRICULTURAL PRODUCTION UNIT 1970 1975 1980 1984 1985 1986 1987 1988 1989 1990 1991 1992 Cereals: Paddy th. ton 1,865 1,813 2,109 2,131 2,060 2,116 2,178 2,449 2,300 2,420 2,342 2,450 Maize th. ton 109 123 128 141 140 153 158 166 160 155 145 150 Sorghum tons 1,920 2,340 1,000 1,500 1.380 1,400 1,485 1,380 1.352 1,352 1400 1.500 Industrial: Sugar cane th. ton 1,239 1,330 1,395 1.660 1,744 1,960 1,990 1,986 1,990 2,000 1,950 1.900 Peanuts ton 41,490 38.300 39,075 31,500 31,015 32,540 32,500 30,200 32,300 30.4 29,000 28,500 Cotton ton 18,606 30,728 23,210 32,600 42,903 40,866 27,000 31.300 41,500 32.100 26,700 27,000 Exports: Green coffee ton 66,556 87,390 79,890 81,400 78,500 78,500 80,600 83,600 88,200 85,000 85,000 86,500 Cocoa ton 1,100 1,300 1.690 3,000 2,300 2,400 2,600 2,700 2,800 3,600 3,700 3,700 Pepper ton 2,116 1,980 2.755 2,610 2,800 2,800 3,000 3,600 3,700 3,600 3,800 4,000 Vanilla ton 1,770 7,680 2,960 6,900 7,000 3,300 7,800 6,800 7,800 8.500 8,000 6,500 Cloves ton 13.430 8,100 12,250 18,000 13,500 7,100 7,100 10.100 7,100 9,600 8,600 8,000 Sisal ton 26,200 20,845 16,020 19.610 19,800 19,800 19,700 19,600 19,900 20,000 20,000 19,000 Roots: Potatoes ton 94,280 171,630 166,160 263,510 263,600 263,900 266,640 270,100 271,000 272,000 274,000 276,000 Cassava th. ton 1,218 1,309 1.683 2,047 2,142 2.198 2,178 2186 2,277 2,292 2,307 2,162 Source: MInsty of Agriculture, 1992 TABLE 2 (continued) D. VALUE OF AGRICULTURAL PRODUCTION (Million of curent FMG) Share of 1992 1984 1985 1988 1987 1088 1089 1990 1991 1992 value of ag. prod. Agrioulture: 773.756 869,024 1,016.613 1,210,640 1,509,598 1.676,755 1.920.498 2,023.787 2,236.843 100% Cereals: 266,777 299,600 329,785 459.315 619,275 683.500 731.497 812.784 946,998 42% Paddy 255,720 288.400 317,400 435,600 590,975 654.500 702,284 784.219 914.046 41% Mae" 11,057 11,200 12,385 23,715 28.300 29,000 29.213 28.565 32,952 1% Industrial Crops: 28,240 32.111 43.844 46.578 111,867 67,235 66.639 69.706 73,689 3% Sugar cane 19,920 20,928 27,300 31,840 39,700 40,795 43,000 45.630 47,500 2% Peanuts 2,520 3.102 4,881 5,850 60,040 8,076 8,269 8.940 9,729 0% Tobacco 795 1,217 1.446 1.787 2.926 3,425 2,800 3.500 3.500 0% Cotton 5.005 6.864 10,217 7,101 9,202 14,940 12,570 11.636 12.960 1% Expc.- Crops: 46,879 60.259 61,103 88,744 104,499 105,819 107,064 129.517 156,138 7% Coffee 26,862 31,008 47,100 64.400 70,975 70.560 68000 85.000 95,150 4% Cocoa 387 340 360 515 562 629 720 851 888 0% Pepper 666 840 840 3,570 7.873 9,260 9.000 12,160 14.600 1% Vanilla 6,900 7,000 3.630 9.360 11.560 13.260 15,644 15.856 32.500 1% Cloves 7,830 5,873 3.728 4.260 6.060 4,260 5,700 7.650 5,400 0% Sisal 4,234 5.198 6,445 6,639 7,469 7.960 8.000 8.000 7.600 0% Roots: 81,840 100,094 135,393 122,892 146.097 161,015 217.392 218.561 224.294 10% Pbtatoes 16,547 16,343 20.684 21.085 25,660 27,913 32,640 35.189 38,364 2% Cassava 51.178 66,645 89.790 76,680 91.825 104,742 148,980 147,648 149,948 7% Others 15.115 18,206 25,019 25,147 28.612 28,360 35.772 35,724 35.982 2% Vegetables: 16,378 21,910 29.309 43,193 48.330 49,788 49.351 54,489 56.788 3% Fruits: 76,542 85,376 94.848 97.472 90.298 112.394 131.555 162,659 160.047 7% Uvstock 28,139 34,796 45,895 46,786 65,156 74,901 82,316 87,748 93,224 4% Cattle 10,16 16,229 19,900 22,920 29,700 34,728 37.613 40.641 44.275 2% Pigs 5.490 8.131 8,297 7,435 8,997 10.055 10,713 11.415 12.390 1% Other 12,033 12,436 17,698 16.411 28,459 30,120 33.990 35,690 36.659 2% ishing: 161.792 163.453 165.649 207.084 280,647 298.388 363,026 402,075 439,211 20% Sea fish 69.618 67.710 75,233 84.753 113,004 134.870 167,961 212.899 247.029 11% Shell fish 14.874 18,463 21.636 29,821 36,519 43,057 68603 72,228 64.603 3% Fresh water 87,300 77,200 88,680 92.620 111.024 120.401 120,461 116,948 127.579 6% Soae.: AUdMpy ot Agdafte, 1992 TABLE 3 A. IMPORTANCE OF IRRIGATION IN MADAGASCAR. BY PROVINCE Ifaritany) Average Faitany Number of farmers % practicing Cultivated area Irrigated aree Total Irrigation Irrigation Total Irrigated % irrigated (he per farm) Antananarivo 331,000 308,000 92% 349.000 214,000 61% 0.70 Fianarantsoa 351.000 299,000 85% 376,000 197,000 52% 0.68 Toamasine 248,000 122,000 49% 357.000 134,000 38% 1.10 Mahajanga 167,000 137,000 82% 271.000 175.000 65% 1.28 Toliara 237,000 116,000 49% 218,000 82,000 38% 0.71 Antuernana 127.000 77,000 61% 184,000 79,000 43% 1.03 TOTAL 1.459,000 1,057,000 72% 1,755,000 891.000 51% 0.84 Source: Ministry of Agriculture, 1992 B. IMPORTANCE OF PADDY IN IRRIGATION AND IN TOTAL CULTIVATED AREA. BY PROVINCE (faritany Area cultivated % of total Cultivated paddy Irrigated paddyl Provincial share of Faritany under paddy cultivated area area under irrigation total paddy national production Antannativo 223.765 65 212.577 95 21 Fianarantsoa 220.366 59 198.329 90 21 Toamasine 216,244 61 129,746 60 20 Mahajange 211,791 57 180,022 85 20 Tollara 83.922 39 81.404 97 8 Antelranans 111.608 61 78,126 70 10 TOTAL 1,067,698 61 880.204 82 100 Source: Ministry of Agriculture, 1992 REGIONAL DIFFERENTIATION IN AGRICULTURAL PRODUCTION 11989) A. Metric Tons Sugar Sweet Butter Peanuts Coffee Cane Cloves Beans Maize Cassava Rice Potatoes Beans Pepper Potatoes Vanilla West Periphery 1.515 172,255 1,285 3,360 60,045 137,210 10.000 2,050 East 22.028 243,365 3,345 225 4.695 122255 141.920 10,805 365 00 Southeast 25 33,628 193.75 150 790 1,945 662,775 1456,030 31,435 1,205 Soutbeast Periphery 85 5.427 11,625 20 285 195 6E,055 29,200 19.485 30 35 Northemet 75 14,837 83,410 3.655 435 2.830 75,945 151,440 91965 216 7.455 Highlands 5.705 15,341 45,015 28,100 58,60 479,185 517,725 249,300 252.175 North Highlands Par. 1.370 1.253 123.155 795 1.475 5,550 168,000 2.975 280 80 Lao Aleotra 1.560 3.399 32,555 15 '80B 7,625 72.755 213,055 3,965 25 740 Northwest 4,780 584 382,070 180 4,880 39,340 257,140 8,280 50 20 Midwest 8,445 213 39,115 2,550 42,220 182,875 279,155 12240 16,765 North 835 5.887 553,886 40 1,810 22,845 107,845 2,905 225 185 South 325 53,825 705 13,110 160.480 22,085 80,020 80 Southwest 3,235 25,300 185 5.470 91,340 46.800 21.716 5,070 Southwest Periphery 4,345 219 51.750 1,640 8.305 203.000 102.195 22.130 1.815 Madagascar 31,825 102,016 1.972,600 7,105 37.080 154,400 2,226,845 2,380.000 483.200 7,280 1,905 271,000 7,800 B. Percent of Production Sugar Sweet Butter Peanuts Coffee Cane Cloves Beans Maize Cassava Paddy Potatoes Beans Pepper Potatoes Vanilla West Periphery 4.8% 0.0% 8.7% 0.0% 3.5% 2.2% 2.7% 6.8% 2.1% 28.2% 0.0% 0.0% 0.0% East 0.0% 21.4% 12.3% 47.1% 0.6% 3.0% 5.5% 80.0% 2.2% 0.0% 19.2% 0.0% 0.8% Southeast 0.1% 32.7% 9.8% 2.1% 2.1% 1.3% 29.9% 0.1% 8.5% 0.0% 6.4% 0.0% 0.0% Southeast Periphery 0.3% 5.3% 0.0% 0.3% 0.8% 0.1% 2.6% 1.2% 4.0% 0.4% 1.8% 0.0% 0.0% Northeast 0.2% 14.4% 3.2% 50.0% 1.2% 1.8% 0.3% 6.4% 2.1% 0.0% 11.3% 0.0% 95.6% Highlands 17.9% 14.9% 2.3% 0.0% 70.6% 30.0% 21.5% 21.8% 51.0% 0.0% 0.0% 93.1% 0.0% North Highlands Peri. 4.3% 1.2% 6.2% 0.0% 2.1% 1.0% 0.2% 7.1% 0.0% 0.0% 0.0% 0.1% 1.0% Leo Alaotra 4.9% 3.3% 1.7% 0.2% 4.9% 4.% 3.3% 9.0% 0.8% 0.0% 1.3% 0.3% 0.0% Northwest 15.0% 0.8% 19.4% 0.0% 0.6% 3.2% 1.8% 10.8% 1.3% 0.7% 0.0% 0.0% 0.3% Midwest 26.5% 0.2% 1.9% 0.0% 8.9% 27.3% 8.2% 11.7% 2.5% 0.0% 0.0% 5.8% 0.0% North 2.6% 5.7% 28.1% 0.0% 0.1% 1.2% 1.0% 4.5% 0.8% 0.0% 0.0% 0.1% 2.4% South 1.0% 0.0% 2.7% 0.0% 1.9% 8.5% 7.2% 1.0% 10.6% 1.1% 0.0% 0.0% 0.0% Southwest 10.2% 0.0% 1.3% 0.0% 0.5% 3.5% 3.7% 2.0% 4.6% 69.6% 0.0% 0.0% 0.0% Southwest Periphery 13.7% 0.2% 2.0% 0.0% 4.4% 5.4% 9.1% 0.8% 4.0% 0.0% 0.0% 0.7% 0.0% Madagascar 100 100 100 100 100 100 100 100 100 100 100 100 100 Souce Regional speelaireto, and sric,uurl grow in Afedagesear. AIRD, 1992 35 TABLë 5 гronaPµ мепМ'снлnкиs� �ап8. �яв4лая2 '.`•гУ::нR:;4:i:�ниг.»�лvн.у.г•i:С:ь:{i?��::г:iг:'+!:•:й ..:. ,. .: :nw::�:г.,.{;.. f{гг°{л.!{г:•{г:• �.> •ч'лМ•м' ':4>.s...f:.a'•}}:'>4:}$:4 : .л.».л. 'гю:м'г::Lа{п..:.лл.н�.. ..:..Sми.л4�глл °+'ь ^�t;.f.+.{L,:M,. h>и}YS...:SY:.:3й.•{ь•.4.• • , . � v�ь::, Ч84 • • 1Ме. ':...'f1M' 18i0` 1М1 • увsааl : Or�V'.. Y+YMf1Y.. fLвrNN.. У+bв/Mi. . �•R ' � � . • Yalrww!М6 YYrвlii0 УвlwвМе м/ . •. � , � УNвм Уiи :: °:.а:;г•г:•>т,.•: :.::::: { :::::.::::::: ,.:. �:. :::гr..{:;'r";� „�',,,.г,1,�„i�'::;�:.5�;�i' 225Д18,2 285Ai8,8 214.7ВВ�{ 110А8/ 298,9®7,9 л<: �:.• . .. ....... ... о1 в�hloh: Cottи 61,337 77А04А 101,051,5 юА�эеА 57Ае5,6 4оА7е s1A02A а -е qow 8,20В 20А85д 21,497,5 51�08д 28А38А 13А78 41АБ3А 11 11 Уапiв 827 30,400д 5ВА11,1 e7,5W�4 85А12,9 842 В4А87,5 -4 16 6'rN 13А88 2А41А м. s,�тТ 1Ае1,2 4,Т58,е s,евод 4мг,s 7,а3о Б,е2вд е 1в Rвffr 1,888 54tA Т,37ед 1,580А 1,6qA 1,t10 2Ав'1А •5 2t wмw ( { о.1 в24А 4sз,2 5ае юz,5. . 8ам 17 44 485�t 1л48,7 2,531,1 9,Т72 9д17А 112 167 Соооо 2�t85 2А23А 7,586.7 3,е11,3 3А`+5.7 4,273 7,274А 8 1е Рв Овг 2А05 2,2е3,8 8А74,5 БА54,Э 9,733,5 1АЧ 4л84,Т -0 10 Мвпбо - - 47д 1�t10A 4А20 В72,9 С'ммюn - - t,050,7 2А05А 1,182� 2А18 Т,Т05А ИИфвlМrЬв 1,200 ец 4,4в7.1 2,7А5,7 2А38,1 787 9,5е2.1 -8 28 ИWев 7,700 4М3 6,28Tw9 6Д91А e,8Z4,2 74,480 3,7ТВ,Т 8 •Т бviв е4 8А 9,557А O,W2A БА18,3 Т,46Т 7,W8A °' ;:. ..° � 5t,5o7A 58,4юр ее,5Т3А и.476 101,и8А ' ��1!l�Т:р1СХС;АWрг1!Мli�1tF';;� е1 wAioИ: FrM1 280 007 55ВА 40еА 1,622,7 788 2,393 � 16 2Т SИвьБ�И • • Тд2Т,5 3,945А 4А18А t,404 е,43е,7 Миt 3ле8 2,54е,.1 1,ОбеА 2,ц0�4 3А7В,7 19А42 tlM0 7 21 14 ем:�. 3,аоо 1ТА14р 42,ъе5,е и,еае,7 воА31,9 е,5ее 74,пТ,s в Те еоае.г. эо 1тs 3,eez,5 э,7зеА 4А15А s1э еАо/А 4о ае г.� i:.^: S;ёл о:<.i;:;>:::;><Е�i1й�;;<»:z:`::.:;i�`:t. 1вл11А 4о,Б3е.Т м,sиА в7,зх+ г5,78в,б �ы.ri�n:° :. . . Coew 28Ае2 6,548А 11,7е7А 9е.564А 30А00,1 21,18В 1В,126,Т -4 18 Сииwд тввt 37е 53еА 3в,2 БА 3е,1 1 37А -57 3Т Имм.» • • 1,�14д о,� 4ео,� 13,�ю 1,13е,1 r.емаииiиме.геьr. . . t31A 284А 180А е1о /At7,2 Lевиа • - 782А е61,е 1351t 2А1Т 474,7 .. <.:iг1[Г11хt�'�.�/Aif;.�i1'1�'„,:;.'•.'' • • 445А 329А 871,7 89 3е8,Т .о1 wИioAo 8и wиг - 40tA 254,7 830А 48 226А ..............тг:'�`�%ч;;;;?<:':'�:� • • { { 115й i9 i/e,1 s%`�tllliti*f�k1`Чtг�1R:;::::?:;'; 33,268,7 53Ае5А 58,589.7 2е9,19е 58,110,3 о1 ыhbA: СИгопkв 82А23 2дее,9 11,217А 31А8�,7 1В,БО7А 128,3W 17,288.1 5 92 hboNиnpeduou 2,Т841 810А ОА>)еА В,061,7 14,983,8 10ВАВ1 20,651,6 74 68 И�иs. е74 2Т7,7 1АБе,Т srs,4 2АоеА и3 В22А •е 2о Grвphicв 14,527 9,565,5 19,982А 16ае4/1 22А59,3 13,555 17А07,9 •1 2е lkоkюryв[в1 БОО 231 185,1 305,Э 2'lе,3 16е 402,1 •16 в s.E 7Аоо 173 ы7А 1,Т18,б еslд 10,3n 1,быд 4 зе :;;�i�N�k�#iqiiiiYf�i;"г; .': 678еА е,з5в,г еА61а 1А56 lвдео,2 е1 whbИ: aewer 1 �е1 3,so4,6 3,687.1 3,s8eA 2А20А е43 4,171А д з rrnFVыper е 1а,7 1,2ад 1anA гsas,7 е 1.1ее,) г 41 .. . ":,::,<:�AYi1�;, ч��,�� i;;'�;; 3,22вА 3.)Т8А 2A77�t sst 4,264,) о1 whlвA: а..ьо.и.r.гмм • 1мА 1,7м,7 1ее,6 Б3Т аеоА wwwaиг 1,зsе 18оа • сгееевr.►ь� • • 371А 7ооА н0А а 2tвд ::.:.:...:.::: ..� ..:� ...., .:..,.:.•:...:,- ���"•� 1Л33,� T.7D�A 2А08,Т 20АН БА01А ....:�#М110.�". ' s:: : � ....... . . .. . .. .... .. ......:;:r,: . . weea • • в0�А 1,ае1А /АеSА а.765 2Ао4А :;;.;��i+R:.;:;;�<�;,° :;:;';, • • Тз,7кА 37А93,2 37А»,б ю,671 eeAw.e ег иА:iвИ: Седм • 5f0.1 2А�А 578,5 3,407 7,р1А бiиllЪив • 9,281,� 4,911.7 7.ОJвА sASO 5,762А e�иuww • • 1,w6,s 2АвоА ав ие,з с.гаs иы+е и�в• wИw1 4,Боо e,6to 18Ае7.е 23�1?2А lвлог,2 4аТе аТАТТА •Т п1 ..,,.. .................. ....>., ::,... . .... .... „�игwмiiг:�.. , ::{>:>::•;:.;:,._ .. ..,....:>.>,:: eм:iвгвeiswetonи • 2,УОе,Т 4,2еТА Б,М1.7 1.128 5,1�7,� тотКги.оRгвгом zag,гsг �вт,2вв,е зes,oeos sое,�еэ,в мо,з�з,? 4е2.вае въs,о�з,з >>в 1 е ЕерМи вww wмNwMy d�pew� b 18N lowa: �ыwt4 Меюг Авимм�t 11Мвк, WaY евМс,1М3 ТАl1Е i l1ЕОиЖА! ЕСnиомlС A1w.nв oF АОвРСи1.Ти1ИL моwlСТlои и мАоА0АrСАсг tиlt NortA NoпAwqt lYlfAwввt NbhYnM l.и АWпв ИопМвп Евв[ 9outM•п 8outh Bouthwввt ONC ррС 01� ORC бf1С DRC DRC ОАС DRC ORC £ Е м п Е п к оь! п п Е п п п оп п Е п �. авD1в fesd Сго�в Рs,нrиrмг од• оsп о.эsь а7вь ass• Psa. о.в �дг о.ьs. �.�е. о.ъ�в �se. lпiОвlвд tnd'1 О.9Е ОА 1.(Fi в i.62 о 0.6У д 0,85 д Ол13 f 1.2 t 7:07 р З•а7 а .rгг�и.диqге..дггвл� о.епд �.�за ose. i,з+. ол� 1 о.в�г о.азр �.аво 1пtNtM kРмп ИквгиМ Олtа д 1.М д 0.17 • О.ав • 0.45 f 0.9В f 091р 1.87 е Мв@в Яiiиfвdtnd�ieмS O.ab 1.1f О.ЕВ 2.а8 О.еВ 1.71 Ф.fi7 7.еВ 0.77 1.74 MMhd lпргелд 0.7t 1.9е Gгiувlвд г 18.Ы q •1з.79 р Сввввл • Явыгhд цвА1г 0.9 0.93 0.2р 0.В7 0Э О.е2 kriвNN 1.01 р 2.$ р vnик.а,иsвiи�нгt►дьWг О.В3д f.вад kгlрвtвд wmntyfвldr 132 д fпqвW 1г�гомдуWдв О.аб д PMhtaвf-lWnf[dtrвdiг 0.9е 1.де , 1YiгRd lnргвмд Oai2 1.39 0.29 0!6 Nripвpd U.id 8 2.�3 д w lвгУУ-ЯвiпW: ОА] t.at � t. МдгrпW Сгидs CвRen-taгqвhиne ОАТ О.а6 OA4i 0.1i! BtaM tвпл ОАе О.� 2.7! g 4.4а р аарг O.t! � �.�{ А 0.41 1.91 } 1.41 J l.У9 i аьмм-�rггг.еwгг о.sэ 1.а 1 ез s.ве a.se •s.+.ч с.игwе.с,о,,: а.ез в.ов Oгоwrdeut Oi О.аВ 09 Еи0нt СгиМ Roбwrв Се1tи • ГгвА7г О.а2 О.l� 0.92 0.39 Упргелд 0.31 0.28 0.� 0.39 0.4t 0.44 Сrаав - Тлд'h О.3е 0,66 uFRвistвd 0.6t ! 0.51 г CwMwHYYshd: 0.72 О.аа fMMaвwЧMNбrt О,3� 1 О.6В Р 0.20 t Олt 1 мгр.аа 7я7 е �з.вв t . e+n.c..k v.ьl,sw.p. м: +.о7 .г.е� аFОгодаг bd 0.47 -Z�6Z пвве е.а 1.3! вг QkвеМидвд и 4е АУеМ:4гев�ивlв риiпвtвr 1: Nвтвktв 6г(ргад воfили: гвhlи вгрвг 1: 4гоипдлиt овkв иtимд to b nропвЪlr М Fbeд гвииNм, лtи fnrrPtnпd k Vвtaпsndry иren: пидWт-ювlв увгМпе �г бгbkви#4: л�и вгрп тг Wr ирдп аг F1F11lЕ: МЬу внивn рг аипвпраlгу; Мгрв+евN ргiмtм k: Свпоn sид оаkв ввгwмд се М го Явiпfвд: втвi fвпп 1: AntstaEв вги.гмдйвгьвиТв Ьwtвг ек АпЬiгlв ir вtвд иlrпп г•w 6N• еоriолюl sиигид W Ов мвЪМ Sвrreв: М1юаl51лев1ьrМм ввlMkrMwal[йвwМlr AГвAвraкw.IlNiD. t831 { Мв/gжr 1PaYвtkn SибSвomrNniгв; NbM/dмk. faбnewy �993 . . ;ABLE 7 1MTЮfNL RICOME А1А1 CON8INMPTION (оомыА 1884 рг(ав1 (BiRan HNG) 7972 1975 i98D 1881 1882 18В3 7984 1985 188В 1887 1В88 18В8 1890 i991 1982 poputпfon 7Л 7.0 8.7 8 8.2 8.4 9J 10 10.3 90.0 11 11.8 11.7 12 12.4 GОРмпаtвtргtавв 17зг 1742 lееэ leez 1е5г зее7 1ев5 17и.е 17ае.1 17ее.9 18ге.а 1юз.5 19аг.э 1ез7 1es�.7 NstlndYaat7oгw 187 189 208 188 18е 1е7 180 192.7 19е.3 196.е 2R5.4 218.8 227.Е 1е4.2 1е7.е GOp at /ctQ aott 15�4 1553 105� 1493 14е8 1478 1505 1521.9 1551.8 1570.9 1829.5 18В8.7 173/.7 1872.8 188В.8 �Rм,mrs 4т8 5г2 5о5 483 sог 515 591 5зе.в 554.э sae.z 58о.е в1о.а в2з.е еге.е еэа.е г,м�.ов г2е 2эе г55 197 1es 171 18г 1ат.8 го5 ги.е 21е.в 2г1.э г18.1 z1е.в 21гв � гп 9.в ае э.е Э.е 9.8 а.е з.е 9.8 �,,, мг� апа w.гrмо � Мwиlвовиiпр 175 193.В 201.2 211 214.8 217.5 2i5.3 213 208.1 8вntow 8+11 789 894 813 795 793 781 787.4 782.5 787.3 824.3 857.5 88г В29.2 810.2 ГoUiooмumption t780 1еВе 1889 1е51 1е55 1е44 1е32 1е8е.5 7е11.7 1е91.2 1822.8 1852.8 1е82 1880.8 18еВ.9 г3wirNporsnnwnt 145 131 175 175 i70 770 182 170.8 188.8 182 147.3 18t.2 179.2 18D.8 174.5 prhвuonrвumptbn 1е44 i555 1765 1478 1485 1474 /470 1S15.e 1471.В 1449.г 1448 UE8.7 15126 1524 148L8 1Мпг of: бОр pr Gpitв tб0 81 87 77 73 72 71 70 е9 88 е8 ВВ 88 В2 е1 3 ACs GOp pr GPRв 100 ТОг Ве 80 81 8i 8/ 80 80 80 7В ВО 78 77 78 ргрrцо (7ом pv С�УКв tep В8 84 71 88 е7 85 е5 81 59 57 58 58__ _ 54 52 Souroe; WoN Bwйt • И„bnN Аотмв Та6'w APPENDIK A REGIONAL COMPARATIVE ADVANTAGE AND PATTERN OF REGIONAL SPECIALIZATION I. CROP ANALYSIS OF REGIONAL ECONOMC COMPARATIVE ADVANTAGES& A. Food Crops 1.1 Rice is always economically competitive for local consumption if produced using traditional methods of production, either under irrigated or rainfed conditions, but not necessarily if produced on the high costs, large-scale irrigation schemes of the west coast (e.g., Marovoay and Samangoky scheme, see Table 3.1). This comparative advantage is further complemented by a vast potential for surplus production in some selected regions: the Lac Alaotra region as demonstrated by the low DRC and the high financial returns to rice intensification; the North-west region which benefits from favorable conditions such as the availability of good land (balboho) under flood recession condition, the potential for intensification in the profitable small irrigated perimeters schemes, and the higher seasonal prices; and the Mid-west because of land availability and natural potential in that region. 1.2 If maize is economically profitable under rainfed conditions for import substitution, in particular in the North, the central highlands, the South and South-west, and the Midwest, it does not appear to hold a comparative advantage for export.21 Potential for increased production is particularly important in the North in view of the technical potential and profitability for intensification, and in the Mid-west as a result of land availability and natural production potential. Despite recent private sector investments to raise yields and increase production in the South-west, there are serious concerns about the sustainability of increased production because of the potential environmental costs (fertility loss and off- site costs) under the prevailing technology. 1.3 Under prevailing low yields, wheat production in the central highlands is not economically competitive with imports either under rainfed and irrigated conditions. Substantial increases in yields could, however, make domestic wheat production economically profitable for local consumption. Barley, currently produced in the central highlands, is economically profitable under rainfed and irrigated conditions. Because the domestic market for malt is limited and exports are not economically competitive, barley production holds little potential as a major component of agricultural development in the highlands of Madagascar. 1.4 Cassava is largely a non-traded commodity throughout Madagascar, but presents a particularly strong comparative advantage, including for export (as a livestock feed), from the midwest and southern regions (south-west and south) under rainfed conditions. This result is consistent with the recent cassava S/ This section summarizes the results of the analysis conducted in greater detail in Regional Specialization & Agricubural &owth in Madagascar, AIRD/MinAgri, April 1993; and MADAG4SCAR Revue Sectorielle de I'lrHgarion, World Bank, Report No. 11504-MAG, February 1993. 2/ 7be fact maize is currently exported from Madagascar to RA-union can be explained by the higher than world price level which prevails in Riunion and the tariff exemption received by Madagascar in the EC market. 39 Appendix A export development from the southern regions. However, similarly to maize production, the long term sustainability of large-scale cassava production in the South-west is being challenged by environmental considerations. 1.5 Potato production is economically competitive with imports in the higher altitude regions of Madagascar: the Mid-west under rainfed conditions, and the central highlands under both rainfed and irrigated conditions. Potato exports even appear to be competitive when produced in the central highlands under improved production techniques. The production potential in the central highlands is particularly important as it is expected to be the principal beneficiary of off-season irrigated land, in view of the scope for intensification (current yields are low) and the potential for land expansion in rainfed areas (hillsides). B. Industrial Crops 1.6 The economic analysis shows a strong comparative advantage in the production of cotton seed cake and even cotton fiber for export when producea under rainfed conditions by small-scale farmers in the South-west region. Similarly, both large estate and smallholder production of flood recession cotton in the North-west region is highly profitable economically both for the local market and for export. Production potential in those two regions is large and stems from land availability and scope for intensification, but remains challenged in the South-west because of environmental concerns. Irrigated cotton production, however, is not economically profitable in the large irrigated schemes of the South- west region (Samangoky scheme). 1.7 Madagascar also exhibits a healthy comparative advantage in sugar production for the local market in the more favorable production regions of the North-west (Namakia) and the North (Ambilobe). Production for export is economically profitable on the US quota market, but not on the free world market. Sugar production is not economically profitable in the less productive East coast region (Brickaville), and on the expensive irrigated scheme of the West coast (Morondava-Siranala).LQ/ 1.8 As a result of favorable climatic conditions, the North-east region presents a strong comparative advantage in the production of oil palm for the domestic market, although not for export, but suffers from exorbitant evacuation costs from this land-locked region. Oil palm production, however, is not economically competitive in the East and South-east because climatic conditions are clearly less than ideal. Coconut plantation for copra produced under a publicly-owned, large-scale condition (Soavoanio) in the North-east is strongly unprofitable economically both for domestic consumption and export, challenging the viability and continuation of the large public investment in this undertaking. 1.9 Peanut production presents a strong comparative advantage for domestic consumption as well as for the export market under flood-recession production technique in the North-west and rainfed condition in the South-west. Production potential is large in both regions because of land availability and the scope for intensification to be achieved through improvements in planting material and animal traction, which the on-going oilseed production project should help capture. The recent boom in edible peanut export (+ 220 per cent in volume terms between 1985 and 1990) is a testimony to Madagascar's production potential and strong comparative advantage for export. 10/ Source: Dossier SDMANALA Rappon de Sythse Final, DGGP, July 1992. 40 Appendix A C. Livestock 1.10 Among livestock activities, dairy production appears to be economically competitive with imported reconstituted milk for urban consumption when produced in the Antsirabe area (Highlands), and economic competitiveness improves significantly when using least-cost mix of feeds (from crop residues and imported feeds from other regions) to meet energy requirements rather than forage-based production. Production potential for dairy is large in the Highlands in view of the scope for intensification (via improved nutrition and animal health, and genetic improvement), the existing dairy processing plants and milk collection network, the apparent dynamism of the nascent dairy cooperative, and the proximity of a large urban market. The combination of these favorable conditions makes dairy production a major potential component of agricultural development in the central highlands. 1.11 Pork production in the Midwest is economically profitable for local and urban consumption, but does not seem to offer a potential for export. It holds considerable production potential given the natural and economic favorable conditions for the local production of livestock feed, together with the proximity of a large market (Antananarivo and Highlands region). D. Export Crops 1.12 Robusta Coffee. The economic analysis shows that robusta coffee production either under traditional or improved production techniques, which is being produced on the east coast (Northeast, East and Southeast regions) as well as the North, is highly competitive even at the currently low world prices, with DRCs ranging from 0.8 to 0.4 in the most productive eastern coast regions. The strength of this comparative advantage may reflect the relative extensive conditions of coffee production in Madagascar that require few inputs. Potential for increased production (from higher yields and increased acreage) is important in all regions but is currently frustrated by the low financial returns to the farm stemming from the low world prices combined with apparently high marketing costs. The financial analysis indicates that, if protection rates are examined upstream by backing-up reference prices to the farm, the Net Protection Coefficient are consistently well below 0.5 which suggest that coffee producers are not receiving a fair share of the export price. High transport costs as a result of the extremely poor transport infrastructure in the coastal regions, but also limited competition at the export level as a result of limited access to and availability of credit, limited connections with world buyers, would to a large extent explain these very high rates of "taxation" of coffee producers. There is ample evidence, however, that this marketing structure is beginning to change as a result of increased competition among traders since the liberalization, which should imply higher prices to producers once world market prices recover. 1.13 These low financial returns on the farm induce coffee producers to allocate more of their labor to hillside rice production for which financial returns per day and per hectare are higher (about 6.000 FMG returns per ha for hillside rice as opposed to negative financial returns per hectare to coffee production)..11/ Research by ODASE suggests that large increases in productivity may be obtained in the Southeast with the use of additional labor for weeding and tree pruning. Labor is currently not being devoted to these tasks because returns are greater in hillside rice production. A scenario to reduce IlI The analysis assumes the full costing of robusta coffee plantation, and helps explaining why coffee producers lack the financial incentives for re-planting coffee. In practice, coffee trees are already planted and actual financial returns are therefore slightly higher, explaining why, despite the low returns, coffee is still being harvested but with little to no labor allocated to improve productivity and quality. 41 Appendix A rural road transport costs by only 20 per cent for coffee beans would more than double net financial returns to production and collection. 1.14 Cocoa. Cocoa production, exclusively produced in the North (Ambanja) appears to be very profitable (financially) and economically, even at today's low world cocoa prices. One reason for this is the premium that Madagascar's cocoa obtains on the world market. As opposed to coffee, returns to labor in cocoa, inclusive of establishment costs, are about 1.700 FMG/day which is well above the daily wage rate. Part of this profitability is due to the high level of subsidization of cocoa seedlings, although availability is an important constraint. 1.15 Cashewnuts. They are currently mainly produced in the Northwest, although there is also production potential in the North. The DRC analysis is based on the actual data from the FAMAMA plantation, a public parastatal. Yields on the plantation are quite low (100-300 kg/ha), the processing equipment and management also quite inefficient. As a result, the crop is at the margin of profitability for export with a DRC of 0.98. Improvements in efficiency and an increase in yields could make this crop one of Madagascar's more attractive exports. The recent export surge as a result of greater involvement from private sector exporters, despite the official export monopoly attributed to FAMAMA, would suggest that cashews exports are quite competitive under more efficient management. 1.! Butter Beans. Butter beans are mainly produced in the South and Southwest and exports have been increasingly steadily over the last few years, although they are still well below the export performance of the 60s and early 70s. Using Rdunion as the potential market, the analysis suggests a strong comparative advantage for exports of butter beans with a DRC of 0.4, when produced under rainfed conditions and a low-input technology in the Southwest and South. Financial returns also appear to be very attractive at the farm level with returns to labor well above the opportunity cost of labor in those regions, and returns to land only exceeded by cotton production in the Southwest. Under those conditions and the existing land availability, expansion of production and exports is a great opportunity. Re-entry into the large British market, however, will require an improvement in the quality of the product and the elimination of the toxicity problem (aflatoxines). HI. REGIONAL COMPARATIVE ADVANTAGE IN IRRIGATED AGRICULTUREI2/ 2.1 Among irrigated crops, rice is by far the predominant crop in Madagascar, being practically the only crop grown in most areas in the rainy season. For this reasou its profitability is the principal issue concerning a strategy for specialization in irrigated areas. A. Areas of Surplus Rice Production 2.2 The two most important areas of surplus irrigated rice production in Madagascar are Lac Alaotra and FIFABE in the Northwest region. Of the two, Lac Alaotra has by far the greatest production potential. Economic profitability is very high there (DRCs are as low as 0.5), especially for more intensive (FN2) production. Furthermore, financial incentives for intensification are strong. Intensification is being delayed, however, for a number of reasons. First, the region is going through 1/ These results are borrowed from AIRD. Etude du Secteur Irrigu6: Etude de I'Economio do I'Irrigation, prepared for the Ministry of Agriculture, June 1991. 42 Appendix A a substantial reorganization with the breakup of SOMALAC, the parastatal that managed the large irrigation perimeter around Lac Alaotra, and the creation of water user associations in its stead. Sedimentation due to upstream watershed erosion is also a major problem. Finally, there is the negative impact on private sector marketing of government intervention in the importation of fertilizer and rice delivered to Antananarivo. 2.3 Production of irrigated rice at FIFABE near Mahajanga is much less profitable than at Lac Alaotra, with DRCs equaling 1.44 for the arriatra (rainy) and 1.04 for the Jeby (dry) seasons. No fertilizer-responsive improved variety is available here, and costs are elevated due to the need for pumping. Subsidies on this scheme are high (EPCs of 1.52 and 1.40) to encourage production. The area has the advantage, however, that it is able to supply rice during the off-season, which helps to stabilize prices in the highlands, and increases the financial viability of production. 2.4 Although the analysis of comparative costs and incentives was undertaken only in the two major surplus zones, it is clear that the region of surplus production in the north extends over a much wider area. Many of the perimeters in this region are much smaller, however, and so are managed at the village or family level. While the economic and financial profitability of these perimeters has not been calculated in the report, there is every indication from the calculations that have been made for similar schemes in the Highlands that production in these smaller perimeters is profitable now, and would in many cases be even more profitable in the future with investment in the rehabilitation of irrigation works and access roads, and the intensification of cultivation. This is a high priority given the contribution that this region can make to supplying the national rice market. B. Highlands 2.5 The Highlands are currently a major rice producing area, but one which is becoming increasingly diversified. DRCs for irrigated rice are uniformly low (0.49 to 0.71) for traditional and improved techniques, suggesting that the region has a strong comparative advantage in growing rice for the local market. Effective protection is relatively modes (0.92 to 1.28), with subsidies on investment in rehabilitation being the major element of that protection. 2.6 In contrast to rice, wheat production in the Highlands is not currently profitable economically, with DRCs of 1.32 and 1.42. It is profitable financially, however, because of high effective protection (1.98 to 2.07), resulting from monopoly profits on wheat imports and high subsidies on inputs used in local production. The lack of economic profitability is due primarily to very low yields (1.5 tons/ha). If yields were increased by 50 percent, the DRC would be well below unity (less than .65). 2.7 As land on the Highlands becomes increasingly scarce and valuable with population growth, rural- urban migration, and expanded industrialization, its use of contre saison (off season) irrigated land for grain production should decrease in favor of higher valued products, such as horticultural crops and dairy products, for which proximity to local markets is important. The DRC for irrigated potato production, as an example, is 0.76 to 0.88, suggesting that the region has a comparative advantage in this crop. Similar results can be expected for even higher valued horticultural crops. 43 Appendix A C. Southeast Coast 2.8 The DRCs for irrigated rice cultivation on the east coast are low and very robust (0.41 to 0.49). In part this may reflect the fact that the import parity price was calculated assume transportation of imported rice overland via Fianarantsoa to compete with rice produced locally Manakara. Sensitivity analysis suggests, however, that even if the imported rice were transport directly from Toamasina via coastal barge, the DRCs would remain quite low. Thus the Southeast coast appears to have a strong comparative advantage in the production of rice for the domestic market. These conclusions should apply generally to the Northeast and East regions of the east coast as well, since production conditions are generally similar, though climatic conditions actually improve as one moves north due to higher temperatures. Nevertheless, efforts to expand irrigated rice production on the east coast are frustrated by poor soils, erratic rainfall, periodic flooding, competition for labor from cash crops. A further local constraint in the east is the lack of suitable improved rice varieties for lowland irrigation; this constraint should not be difficult to ov*rcome given availability of improved varieties for similar climates in other countries. D. West Coast 2.9 The river valleys of the west of Madagascar appear to offer the greatest technical potent for expanded irrigation production. Land constraints are not as severe as in other parts of country, and water is adequate to support a large expansion of irrigated production. Outside of FIFABE project, however, relatively little has been done to develop this region. Costs of irrigation are high, basic transport infrastructure is lacking, and there are major problems of sedimentation resulting from erosion in the watershed areas. SAMANGOKY, the largest of the west coast schemes near Toliara, has also experienced problems of ethnic conflict resulting from the need for immigration from more populated areas of the country. Lastly, land use and yield performance have been low relative to expectation in all the irrigated schemes of the west, due in large part to technical and social problems and ineffective management by the large parastatals that operate these schemes. 2.10 The economic and financial analyses for SAMANGOKY suggest that extension of irrigate area under existing conditions of parastatal management could be financially profitable because extensive input subsidies, but there would be large economic losses unless a high degree intensification for rice were attained (DRC equal to 1.66 for low input use, 0.93 and 0.81 for high and highest levels of input use). Irrigated cultivation of cotton, maize, and butter beans would even more economically unprofitable than rice (DRC equal to 2.02 to 18.61) and would have negative financial returns. Only irrigated cassava production appears in the analysis to be almost profit (DRC equal to 1.01) for import substitution. Lastly, under a scenario of smaller-scale private investment and management, positive economic returns do appear feasible for irrigation development in the western valleys, although these scenarios imply a much more gradual rate of development theses regions than do the public schemes. 44 Appendix A III. PATrERN OF REGIONAL SPECIALIZATION3/ 3.1 Overall, results of the irrigation and rainfed crop analyses of comparative advantage suggest that the coastal areas have a strong advantage in production of cash crops for export, whereas the Highlands and the Midwest should concentrate on production for the local market. 3.2 Among the rainfed crops that can be profitably exported from the coastal regions are coffee, cocoa, cotton, edible peanuts, cashews, rice, cassava, butter beans, and peanut oil. Although maize is exported, this appears to be due more to European trade policy than to underlying comparative advantage. in fact, the possibility exists for Madagascar to mix feed involving maize and other inputs and to export this directly to Europe or elsewhere, rather than passing through Mauritius and Rdunion. 3.3 Among the non-irrigated agricultural products produced in the Highlands or the Midwest, only potatoes, pork, and possibly arabica coffee (under highly optimistic assumption) appear to have a potential comparative advantage for export. On the other hand. the list of products that can profitably be substituted for imports includes rice, maize, potatoes, wheat, triticale, barley, milk, arabica coffee, pork, and tomato paste. The profitability of food crops such as rainfed rice may be diminishing in the Highlands as the opportunity cost of land rises. This is much less true in the Midwest because of lower population density and greater land availability. 3.4 As far as regional comparative advantage within Madagascar is concerned, the Highlands has a strong comparative advantage in producing rice, maize, potatoes, and milk for its own urban markets. Its comparative advantage in wheat, triticale, barley, and arabica coffee is less pronounced, and these crops should therefore be less favored when they compete for scarce land and labor with crops that are more profitable. The Midwest has a strong comparative advantage in the production of rice, maize, potatoes, cassava, pork, and tomato paste for shipment to the Highlands as well as for local consumption. It may even be able to export cassava if marketing costs can be reduced. 3.5 Of the coastal areas, the North has a greater comparative advantage in producing rice for the local urban market than for export either to Antananarivo or to East Africa. The Northwest has a greater comparative advantage in the production of rice for either local consumption or shipment to the Highlands than for export, whereas its comparative advantage in cotton, though profitable for shipment to the Highlands, is less so than for export. In addition to its obvious natural advantage in producing vanilla and cloves, the Northeast shows strong potential for exporting robusta coffee and selling palm oil domestically. Domestic coconut production does not look attractive, however, in light of low world prices. 3.6 The comparative advantage of the Southwest in producing maize, cassava, butter beans, peanut oil, and cotton is strong for the local market at Toliara but the size of this market is an important constraint. Exports of these crops appear feasible, although the reduction of export costs would greatly improve this potential. The South shows little capacity to produce surplus rainfed food crops efficiently with the exception of cassava, which it may be able to export if the costs of exporting could be lowered. Finally, the Southeast can produce rice economically for sale in Fianarantsoa but not for export. It is currently a deficit rice production area. 131 This section borrows from Regional Specialization and Agricftural Growth in Madagascar, AIRD/MinAgri, April 1993. 45 Appendix A 3.7 Finally, It is important to note that the analyses above are an Incomplete representation of agricultural potential in that they exclude a number of agricultural activities that are found in Madagascar. Examples include tropical fruits, temperate fruits, fresh and frozen vegetables, fois gras, tobacco, and a number of other high-value crops. In general, these products represent very small markets domestically and so are not of major concern in examining food strategy. For export, on the other hand, these products hold very important potential. Often, they face narrow, high quality, or specialized international markets for which marketing strategy and quality control play as important a role as factor costs in determining competitiveness. Despite these omissions, broad patterns of regional comparative advantage have been identified in this chapter. They will serve as the basis for the projections of regional demand and supply found in the next Annex.