I. PRESENT SITUATION OF AGRICULTURE

7. Despite its importance to the economy, agriculture in LDCs has remained largely underdeveloped in production both for the domestic market and for export. Although there was a modest growth of output during 1995-98, it barely exceeded population growth, and for the 1990s as a whole in per caput terms it actually declined. In addition, slow food production growth and sharp annual fluctuations in output remain major and chronic problems for the LDCs, constituting the major causes of their rising poverty and food insecurity. Between 1969-71 and 1996-98, the proportion of undernourished in total population in LDCs increased from 38 percent to 40 per cent, while the absolute number of undernourished increased from 116 million to 235 million. For the rest of the developing countries, by contrast, the proportion of undernourished in total population in 1996-98 was18 percent. In addition, indicators of poverty show that the proportion of people living below the poverty line (defined as $1 per day) has risen in many LDCs.¹ What follows is a brief analysis of the major internal factors underlying the present agricultural situation in LDCs. Areas are highlighted where improvements in policies, institutions and investment could accelerate agricultural growth to levels that would help to reduce rural poverty and enhance food security.

A. SUPPLY ISSUES

1. Trends in production

8. Over the past decade, agricultural production, including food production, has not kept pace with population growth in LDCs as a whole. Although agricultural output in 1990-99 rose at an annual average rate of 2.5 percent, exceeding the rate of 1.6 percent in the previous decade, in per caput terms there was virtually no increase in output, or even a slight decline. The situation was the same for per caput staple food production (Table 1).

9. However, these aggregate figures conceal a wide diversity of performance among countries. While more than 25 countries experienced negative per caput growth rates during 1990-99, 5 had positive growth as high as 2-5 percent. In only about 15 LDCs was per caput agricultural production in 1990-99 higher than in 1980-90. Elsewhere, mainly in SSA, there was a decline.

10. Many LDCs changed from being net food exporters during the 1960s to net food importers during the 1980s and 1990s.Current projections are for their dependence on imports to increase at least up to 2015.²

11. Although there have been sharp annual fluctuations over the past 30 years, the value of production of nearly all agricultural commodities rose during 1990-97, the only exceptions being cassava, cocoa and sisal (Annex Table 1).

Table 1. Agricultural and food production in the 1980s and 1990s in LDCs and other developing countries (Annual average percentage increase)

Source: Computations based on volume indices (FAOSTAT 2000).

2. Determining factors and constraints

2.1 Physical aspects

12. Most LDCs have considerable unexploited potential in agriculture, thanks to their factor endowment in land, water, climate, the scope for utilizing their human resources and improving on their so far limited use of modern farming methods. There is thus great scope for more effective use of their agricultural resources and for increasing their agricultural productivity.

2.1.1 Land and water resource potential and constraints

13. The most fundamental factor influencing the agricultural production potential of a country is the availability of arable land. Land is the essential prior resource needed for crop, animal and forestry production. Thus, the existence of a potential for expanding the cultivated area is basic to national agricultural planning. Comparing the potentially cultivable area with current use of land and forecasts of future population growth will indicate whether countries have the physical capacity for expanding agricultural production, whether for domestic use or for export.

14. Least developed countries have widely diverse agroecological situations, with varying availability and quality of arable land and varying climatic conditions. Some countries have large areas of arable land and considerable water resources while others have more limited availabilities or are almost devoid of these resources. Prospects for agricultural development necessarily hinge on these considerations.

15. With the objective of classifying countries in terms of potential for agricultural production, a ranking on the basis of the land resource availability and constraints was undertaken, taking into account not only land and water constraints but also climatic constraints and population growth.³ Annex Table 2 ranks 35 of the LDCs, for which comprehensive data were available, in terms of per caput potential arable land as well as per caput cultivable land in actual use. Potential arable land refers to areas that could be brought under cultivation because of soil suitability and availability of water (rainfall or irrigation).⁴ It is important to note that in most cases potential arable land is rainfed and suffers from constraints such as ecological fragility, low fertility, toxicity, and high incidence of disease. These reduce its productivity and require heavy inputs and management skills to permit its sustainable use. Furthermore, especially considering the lack of financial resources in many LDCs, prohibitively high investments may be required before the land is rendered accessible or disease-free. FAO projections to 2015 indicate that the expansion of arable land as well as harvested land is expected to be below the past rate of increase.

16. The overall rankings indicate countries with the most favourable conditions (low rank numbers) or the most severe problems (high rank numbers) with respect to physical resource potential and constraints, now and in the future. This ranking is broadly indicative of a country's relative land resource potential. Three types of countries can be distinguished: i) those with a relatively large land balance, where extensive agricultural expansion may still be possible (e.g. Democratic Republic of the Congo and Mozambique); ii) those which are close to the limit of exploiting actual arable land (e.g. Bangladesh and Somalia); and iii) those which have exploited almost all their arable land and can probably not expand much more (e.g. Afghanistan and Yemen). Thus grouped, the countries can respectively be considered as having a high, medium and low agricultural potential. Out of the 10 highest-ranked countries 8 fall in the humid zone of central Africa. In this group there would appear to be a productive potential that is not yet exploited.

17. Among the lowest-ranked countries, there are two highly contrasted groups: i) two countries that have over 90 percent of their land as deserts and drylands; and ii) four relatively humid countries with problems of steeplands and land degradation.

18. Another feature of the lower-ranking countries that may be noted is that at least five of them have, in recent years, experienced major civil conflicts, political instability, or war. The high rate of population growth in these countries is likely to increase pressure on land resources, which can lead to the breakdown of traditional property rights to land, and ultimately of law and order. Among the many consequences of such changes is further degradation of land.

2.1.2 Potential for growth in agricultural productivity and its importance

19. In LDCs, the contribution of increases in productivity to agricultural growth has been limited or zero. Horizontal expansion, i.e. bringing more land under cultivation, remains the dominant source of growth. Given the increasing pressure on agricultural resources, however, faster agricultural growth, particularly in countries with limited scope for land expansion, will require continuing increases in agricultural productivity from its present relative low level. Such increases are attainable if major constraints on enhancing productivity, such as lack of favourable incentives, limited rural public investment and poor institutional support, are effectively tackled.

20. Available evidence shows that the potential productivity gains are considerable. In terms of agricultural value added per worker, productivity increased, though only slightly, in 21 out of the 31 LDCs for which data are available between 1979-81 and 1995-97 (Annex Table 3). However, in comparison to other developing countries the agricultural value added per worker in LDCs appears to be relatively low, suggesting that there is much room for improvement.

21. The following is a broad assessment of productivity in each of the major agricultural sub-sectors.

Crops

22. The most widely used indicator of crop productivity is production per unit of land (also referred to as crop yield). In general, crop yields in LDCs are low relative to those in other developing countries (Annex Table 4). Yields of the basic food commodities (cereals, roots and tubers and oil crops) are less than half the average for developing countries, although there is much variation among countries. There thus appears to be potential for substantial gains in productivity.⁵

23. Unlike most other developing countries, growth in agriculture in LDCs owes a great deal to area expansion rather than to advances in yields. For example, area expansion accounted for 77 percent of the growth in cereal production in LDCs during 1981-89 and for 72 percent in 1990-99, and higher yields for only 23 percent and 27 percent, respectively (Annex Table 5). For rice, maize and fibre crops, however, a relatively high and increasing contribution was made by productivity improvements (yields).

Livestock

24. Livestock is an important and growing sub-sector, providing a substantial source of income and nutrition for the rural poor in most LDCs. It remains the principal form of non-human power available to rural farmers, and is used by both men and women for various purposes, including accumulation.

25. LDCs have substantial hidden growth reserves in the livestock sector. A comparison of LDCs' share in world livestock numbers with their share in world output therefrom (Annex Table 6), provides an indication of the relative productivity levels of LDCs. Although 14 percent of the world's cattle and 18 percent of the world's sheep and goats were in LDCs in 1997-99, those countries produced only about 4 percent of the world's beef and 11 percent of the world's sheep and goat meat.

26. Livestock production in LDCs relies much more on traditional operations. It relies largely on growth in the number of animals for increased production. There was virtually no significant improvement in productivity per animal in most LDCs, where their average productivity levels remain much below those of developing countries as a whole (Annex Table 6). The extreme scarcity of capital, shortage of quality feed and widespread prevalence of disease have constrained their livestock sector. To achieve greater improvements in productivity there is a need for: i) continued investment in both research and the development of animal and feed grain production and processing and ii) assistance to small, poor livestock producers, so that they can become better integrated with commercial livestock marketing and processing.

Fisheries

27. Many LDCs have great potential in fisheries. Although this potential has not yet been fully exploited, fisheries products are increasingly contributing to food consumption of the population and to foreign exchange earnings. Catch potentials vary widely among countries. Those in north-west Africa, south-west Africa, the south-west Indian Ocean (for tuna) and the Rift Valley lakes, for example, have the greatest potential for production and exports. The countries bordering the Atlantic Ocean benefit from particular oceanic conditions (i.e. upwelling systems) that greatly contribute to the increase of marine water productivity, although these systems are subject to marked fluctuations due to weather. High-price demersal species are considered to have approached the limits of possible exploitation, but low price pelagics are thought to be largely under-exploited. Countries still depend largely on foreign investment or international fishing agreements for the exploitation of their offshore resources.

Forestry and agroforestry

28. Forests and trees indirectly contribute to economic development and food security and sustainable livelihoods in numerous ways, through support to agricultural systems, their role in rural development and in maintaining environmental integrity and the provision of opportunities for income generation and employment.

29. Rural communities, particularly in LDCs, are highly dependent on forest goods. Wood fuel is the main source of energy in most LDCs, representing up to 90-95 percent of domestic energy consumption. Non-wood forest products (NWFP) are of major significance primarily in households and local economies. An estimated 80 percent of the population in the developing world use NWFPs to meet some of their health and nutritional needs. Millions of households depend heavily on these products for subsistence consumption and/or income. Timber and the timber industry are an important source of income and a significant component of the national economy in LDCs with high forest cover, representing in one case 15 percent of GDP and 35 percent of total export revenue.

30. In LDCs, forest goods and environmental services are provided almost exclusively by natural ecosystems, which are threatened by unsustainable exploitation practices and other factors, including inappropriate horizontal expansion of crop production. The challenge will be to define and develop integrated systems that ensure sustainable provision by forests and trees of goods and services which are vital to the livelihoods of the population in LDCs.

31. The foregoing analysis shows that the gap between actual productivity levels (in terms of land, labour or animal head) in LDCs and what is potentially achievable is huge. This gap can be defined at three levels:

• Average productivity is far below what could be achieved by using the best practices and technology suitable for the specific location. There is scope for closing this gap through extension programmes and infrastructure investments;
• Yet further improvements in productivity could be obtained through more applied research programmes. However, this involves a long time span and requires continuity of support. The experience of maize in SSA shows that African countries are slow in acquiring the capacity for developing reliable and cost-effective systems for the delivery of crops from the laboratory to the field;⁶
• The difference between productivity derived from scientific innovation and from research is also high. To close this gap applied adaptive research programmes must be supported by international and national pre-invention science programmes.

32. Improving agricultural productivity is associated with the progressive reduction of each of these gaps - starting with the extension gap, moving on to the research gap and then to the science gap - as the country's capacity expands for adopting and developing improved technologies. At this stage of development in the LDCs reducing the extension and research gaps would be the immediate priorities.

33. In many developing countries (including LDCs), governments have often intervened in markets in inappropriate ways and have invested in state-owned production enterprises that have often been inefficient. Reforms have been undertaken to privatise inefficient state-owned enterprises and to eliminate marketing boards and other inefficient regulatory agencies in many countries in recent decades. However, the historical role of such institutions and the associated provision of these public goods in agriculture has not always been fully appreciated. Public sector investment in rural schools, in the development of input and output markets, in agricultural extension and in applied agricultural research have been vital to agricultural development in every economy in the world. Institutional reform without investment in these public goods does not produce economic growth in the agricultural sector. Growth is not produced by passive "let the markets work" policies that do not include critical public investment programmes.

34. Evidence shows that public spending on agricultural extension and research has a potentially high payoff in LDCs. A recent overview of studies on returns to investments in research and extension confirmed that the internal rate of return in Africa (which contains the largest number of LDCs) is rewarding: the median return was 27 percent for extension and 37 percent for research (Annex Table 7). Therefore, building extension and research capacity is necessary to enable LDCs to achieve high productivity growth, in line with the experience of many developing countries. As the programmes are complemented by institutional investments in markets and infrastructure, their effectiveness increases.

35. Despite their high potential payoff, agricultural research and extension expenditures in almost all LDCs are very low in per caput terms compared with those in other developing countries or the developed countries. A recent study by FAO has shown that, in 1989-90, total expenditures on agricultural research in SSA countries were less than 0.6 percent of agricultural GDP.⁷

2.1.3 Environmental and natural resource sustainability

36. There is a growing concern that the expansion and intensification of agriculture may lead to degradation of the natural resource base (soil, water, vegetation and biodiversity) and consequently to a decrease in agricultural production. However, agricultural intensification per se - i.e. increasing the productivity of land already under cultivation - should not be a threat. In fact, properly managed intensification is needed to meet agricultural production needs and reduce the pressure of agricultural expansion in fragile and marginal areas. The lack of sound management practices and of access to appropriate technology and inputs for agriculture, rather than intensification, is the most serious cause of environmental degradation.

37. Sustainability of environmental and natural resources in LDCs is related to a number of factors, such as globalization, inequalities in the development process, lack of access to science and technology, limited financial means of production and disrupted traditional institutions and production systems. Moreover, agriculture still functions as an isolated sector in many LDCs. Increased stress on natural resources, encroachment on marginal lands, migration towards cities or more developed countries, urban slums, social disintegration and poverty often result from a lack of services and employment opportunities in rural areas in the LDCs.

38. Population growth and unsustainable management practices create pressures on the eco-system and jeopardize the ecological balance. The last five decades of resource over-exploitation in many LDCs have drained reserves of natural capital in many regions and limited agricultural and livelihood opportunities for future growth. Revitalization of indigenous knowledge and more research are needed on production methods that preserve natural resources and the environment. Most importantly, more attention should be given to local participation in decision-making processes for better interactions of individuals and social groups with the natural ecosystem. There should be financial and other assistance to help LDCs adopt and acquire appropriate technology. Programmes, including a diagnosis, at national, subnational and local level, of areas and populations most exposed to the degradation of land resources, and with consequences for their livelihood, need to be initiated and implemented jointly by various stakeholders. Unfortunately, these areas are often left aside by development programmes, as their rehabilitation is complex and not easy to justify on purely economic grounds. The selection of appropriate inputs should also be based on solutions that combine traditional knowledge and modern techniques and assist the farmers in investing in the maintenance of land assets.

39. The integration of environmental considerations into development planning should be seen as an indispensable element of development strategy in LDCs. It is not only a means of protecting fragile lands for future agricultural production, but also a mechanism for LDCs, in particular those in dryland zones, to sustain an important capital of biological diversity and contribute to solutions to some of the global change issues. Thus, a site-by-site analysis is likely to be required, given that the interaction between policies that promote a supply response and the manner in which that response will be achieved (and hence the environmental impact) is likely to be ambiguous.

40. In sum, it appears that many of the LDCs have relatively abundant agricultural and natural resources that could provide them with a comparative advantage in a range of agricultural products. These could be developed to exploit international market opportunities and therefore generate broad-based growth throughout the economy. There are great opportunities for intensification and productivity enhancement in agriculture. The next three sub-sections examine domestic and external challenges and constraints that have impeded the full exploitation of this potential and highlight policy measures for its realization in an effective and sustainable manner.

2.2 Human development aspects

41. Developing the human resource potential involves examining the roles and needs of farmers (both men and women) and other members of the household who may perform diverse duties and have differing requirements with regard to education, health and nutrition, and technical knowledge. A low level of human development (as measured by a combination of life expectancy rates, education attainment rates and standards of living)⁸ is characteristic of LDCs.

2.2.1 Education, training and extension

42. Education is the main pillar of human development and a major factor in agricultural development. Research shows that primary education attainments and literacy, training in basic skills and extension services have an immediate and positive impact on farmers' productivity. A farmer with four years of elementary education is, on average, 8.7 percent more productive than one with no education. Moreover, the better he is educated, the more he stands to gain in income from the use of new technologies and the more rapidly he adjusts to technological changes. The effects are beneficial to the whole population ; more specifically, they enhance the capacity of the rural population.

43. The quality of education and training in LDCs is low, and the institutional capacity to carry out reforms and improvements in education and training for agriculture and rural development is weak. As a result, LDCs have high rates of illiteracy and of children out of school, affecting most acutely the rural population.

2.2.2 Population and health

44. Demographically speaking, the LDCs suffer from a dangerous combination of population, health and development problems that add up to a daunting challenge for their people, their governments and the international community.

45. Current projections indicate that they will continue to experience a high national rate of population growth, although it could be set back by the AIDS epidemic, if unchecked. Obviously, the projected increases in overall population numbers will have major implications for food requirements. For instance, a recent FAO study ⁹ indicated that in order to maintain, or slightly improve, present per capita food availability by 2050, food supply would need to be nearly quadrupled in some LDCs.

46. There are other demographic factors that are likely to be of direct relevance to agriculture and food security in LDCs. In particular, the increasingly rapid spread of HIV/AIDS in rural areas poses a very serious problem. The pandemic is unique in comparison with other diseases in that it affects the most productive age groups: those between 15 and 50 years. It thus has direct quantitative and qualitative effects on agricultural labour: it greatly reduces the size of the agricultural workforce and its productivity; it changes the division of labour; and it results in a loss of skills that are important for farming, marketing and management of resources.¹⁰

47. The pandemic also directly affects markets for agricultural production by altering the size and composition of the population to be fed, and limiting its effective demand for food. In addition to being a major health problem, in recent years HIV/AIDS has been considered a critical socio-economic issue. Its impact is also related to the fact that it provides an entry point for other diseases, such as tuberculosis and malaria. Increased levels of morbidity and mortality impoverish affected households and deplete the rural sector at large. This is likely to lead to declines in agricultural production and to aggravate food shortages and long-term nutritional deficiencies.

2.2.3 The role of rural women in agricultural development

48. Rural women play an important role in producing the world's staple crops, raising poultry and small animals (sheep, goats, rabbits and guinea pigs), and providing labour for post-harvest activities. Their role is particularly prominent in LDCs. Wars, increasing rural-to-urban migration of men in search of paid employment, together with rising mortality attributed to HIV/AIDS, have led to an increase in the number of female-headed households in the developing world. This 'feminisation of agriculture' has placed a considerable burden on women's capacity to produce, provide, and prepare food in the face of already considerable obstacles.

49. FAO studies demonstrate that while women in most developing countries are the mainstay of agricultural sectors, the farm labour force and food systems (and day-to-day family subsistence), they have been the last to benefit from - or in some cases have been negatively affected by - prevailing economic growth and development processes. Gender bias and blindness persist: farmers are still generally perceived as 'male' by policy-makers, development planners and providers of agricultural services. Women consequently find it more difficult than men to gain access to valuable resources such as land, credit and agricultural inputs, technology, extension services, training and other services that would enhance their productive capacity.

50. Overall, women's contribution to agriculture is poorly understood and their specific needs ignored in development planning. However, women's full potential in agriculture must be realized if the goal of promoting agricultural and rural development is to be achieved.

2.2.4 Information and communications

51. Information and communications are also essential for sustainable agricultural and rural development. Investments in rural information systems can improve farmers' knowledge levels and management skills. Raising the level of awareness, acquiring information, sharing experiences, changing attitudes and developing skills call for processes of communication and learning. While Internet-based technologies are spreading rapidly in many developing countries, there is still a serious lack of basic telecommunications infrastructure. The information gap between the rich and the poor is indeed very wide.

2.3 Policies and institutions

52. This sub-section identifies the major policy and institutional measures that have facilitated or constrained agricultural development in LDCs, with emphasis on those that prevented farmers from increasing their productivity or output.

2.3.1 Macroeconomic policy framework

53. In the past, governments in LDCs used to carry out many of the functions associated with agriculture: funding, a variety of production, marketing and distribution services, regulation and in some cases direct involvement in production. More importantly, the overall effect of government policies was not favourable to the agricultural sector. The prevailing development paradigm emphasized the importance of extraction of agricultural surplus in favour of other sectors. Macroeconomic policies, especially exchange rate policies, discriminated against tradables, while trade policies, by favouring non-agricultural tradables, "tilted" the terms of trade within the tradable sector against agriculture. More importantly, though, the price-based bias against the agricultural sector was not compensated by other forms of transfers in favour of rural areas.

54. Since the early 1980s, most LDCs, like many other developing countries, have been implementing a series reforms both to address macroeconomic disequilibria and to rectify the distorted inter and intra-sectoral price incentives. At the macroeconomic level and in the context of stabilization programmes, a major change has been a move towards an exchange rate system better reflecting the scarcity of foreign exchange and a monetary and fiscal policy conducive to macroeconomic stability. Thus, a major source of anti-agricultural bias has been addressed but not necessarily entirely removed. And steps towards macroeconomic reform have not been uniform in all countries. At the sectoral level, steps have been taken to remove distortionary barriers to the functioning of markets, and towards privatization of processing, marketing and distribution activities.

55. The diversity in the contents and in the implementation of reform "packages" makes it impossible to undertake an overall evaluation of their impacts on agriculture (by e.g. comparing pre-and post-adjustment growth in agriculture or countries which adjusted and those that did not). With respect to macroeconomic policies, it is nevertheless relevant to note that in a number of LDCs which experienced buoyant agricultural growth, macroeconomic policies brought about an increasingly competitive exchange rate and more realistic interest rates.¹¹ A stable macroeconomy, by promoting investor confidence, constitutes an essential characteristic of an overall growth environment, which in turn induces an expansion of the internal market for agricultural commodities. A competitive exchange rate promotes agricultural exports.

56. As for sectoral policies, the limited cross-country evidence on the impact of sectoral policy reform on agriculture has shown that, while improving the structure of price incentives facing agricultural producers is important, it does not address all the constraints which prevent agriculture from realizing its productive potential. In fact, examination of a number of successful and sustained agricultural growth experiences shows that, in certain periods, output growth has taken place even though the structure of price incentives was not favourable to agriculture.¹² The principal difference between these success stories and others, where price discrimination against agriculture resulted in the stagnation of the sector, is that, despite price distortions, there was nevertheless a government (and donor) commitment to building rural infrastructure and promoting agricultural research and other public services, which more than compensated for the loss caused by distorted price incentives.¹³

57. Thus, the major lesson that emerges from country experiences is that for agricultural growth to occur, a number of factors need to be in place which address the "handicap" of the rural sector in terms of infrastructure, social services, technology, marketing infrastructure, and seasonal credit availability, along with the building of an appropriate institutional environment. There is no unique policy prescription that fits the diversity of the agricultural sector in the LDCs. While enhancing productivity is a common essential requirement, the nature of the increase in productivity envisaged will determine the appropriate policy mix. For example, in countries seeking increased productivity through shifts to commodities with a higher income elasticity of demand (such as fruits and vegetables) and through improved access to dynamic markets (both domestic and external), an appropriate institutional environment, market information and assistance in meeting health and sanitation standards are some of the possible elements of policy.

58. In addition, the multiplicity of linkages of agriculture to the broader rural sector and rural non-farm activities suggests that agricultural policy should not be confined to the narrow limits of the agricultural sector strictly defined, but should consider also the impact of policy on the rural space for which agriculture (especially in LDCs) is the central activity (see section C below).

59. In several countries reforms have not been properly sequenced so as to ensure their efficiency. In designing policies and programmes governments have often concentrated on exchange rates, domestic price liberalisation and privatization of public enterprises, while downplaying other policies and factors affecting agriculture, such as an accompanying adequate improvement in infrastructure, technology and marketing facilities. Inadequate design and sequencing of reforms, and an unstable policy environment, have thus been major sources of the difficulties faced by the reform programmes in many LDCs.

2.3.2 Agricultural and rural development institutions, infrastructure and support services

(i) Markets

60. Rural infrastructure in most LDCs is rudimentary, with semi-subsistence farming often dominating agricultural activities. Lack of or difficult access to markets is common to most LDCs. Even where rural markets exist they are notoriously imperfect, and when they are totally absent it is difficult for farmers to sell their produce and thus ensure food security for their families. An initial requirement is frequently thus the development of these rural markets. Difficulties that have been cited in the operation of commodity markets include remoteness of producers from markets, poor quality of the produce, high transport costs (because of high energy prices and weak infrastructure), lack of competition among traders and poor organization of producers, lack of information on market conditions, lack of clear market rules and their poor enforcement, as well as sharp price fluctuations during the year.

(ii) Rural financial services

61. Financial services in rural areas are often poorly developed. The channelling of cheap credit through state agricultural development banks was characterized by low repayment rates, poor targeting and low operational and managerial efficiency and thus was limited in terms of outreach and sustainability. Often subsidised credit has been misused and channelled towards the introduction of technological packages that were not adapted to local farming systems and for which no effective demand existed. Poor assessment of marketing possibilities and profitability and the limited loan repayment capacities of the borrowers often explain the high rate of loan defaults, reinforced by periodic debt waivers advocated through political pressure.

62. In contrast, private commercial banks charge high interest rates, especially to small farmers in regions with low population densities. The consequently high costs of borrowing are further increased by an unstable macroeconomic environment involving, inter alia, high annual inflation rates. In addition, poor rural infrastructure and communication systems, ineffective extension services, and inappropriate macroeconomic and sectoral policies raise the costs of inputs and marketing, further reducing the profitability of farming. Linkages between farmer and trader and other arrangements with enterprises in the agribusiness chain, such as contract farming, can overcome many of these constraints.

63. Experiences with microfinance institutions highlight the crucial importance of client orientation in the provision of financial services and the use of market- based interest rates that cover the full costs of lending. Poor people seem to prefer a reliable and timely availability of loan finance, even at higher costs, to an untimely and bureaucratic supply of subsidised credit that is tied to specific uses.

64. However, the specific nature of agriculture, such as seasonal credit demand for annual crops and high risks, reduces the role of current microfinance institutions and their lending methods in financing the seasonal and on-farm investment needs of small farmers.

(iii) Availability of farm inputs

65. Information gathered through FAO's Special Programme for Food Security (SPFS) projects in 22 of the low-income food-deficit countries (LIFDCs) shows that a major problem facing farmers is the unavailability of fertilisers and agro-chemicals, and often of animal feed, on time or in the quantity required. This constraint is largely linked to the lack of credit, difficulties in obtaining foreign exchange, the seasonality of agricultural input requirements, spatial dispersion of farmers, poor transport infrastructure and, sometimes, to the marketing and management inefficiencies of the state-owned companies responsible for single-channel input supply and marketing.

66. Quality seeds are also said to be available in insufficient quantities, particularly in Africa and Asia. The informal seed supply system is the dominant source of seed/planting materials for resource-poor farmers in marginal areas and has proven to cope better with a disaster situation compared to the formal seed sector. Nevertheless, the informal seed supply sector has unfortunately received very little attention and financial support from policy makers, to the detriment of the productivity of small-scale farmers. Therefore, without strengthening seed supply systems in developing countries there will be little or no technology transfer to improve crop productivity and hence the livelihoods and well being of poor and vulnerable households in rural communities. In some countries, there are worries that the genetic base of certain cereals has become too narrow, especially as local varieties have been given less importance or suppressed. Absence of improved animal breeds and insufficient livestock treatment facilities are also reported in some cases.

67. Another institutional constraint is inefficient use and distribution of water, which is usually blamed on poor management of irrigation schemes and inadequate water distribution arrangements, which result in an uneven and untimely distribution of water among farmers. In many LDCs, the management of irrigation schemes and water distribution is under public control. Farmers' associations are rarely involved or are too weak to contribute to both the design of water distribution systems and the maintenance of the network. The water needs of farmers have to be examined from both the household and production-for-export aspects, since the particular use affects the quality of life of both men and women and their communities.

(iv) Agricultural research and extension

68. In most LDCs, the institutional capacity for research and extension is weak. As a result, the technology available is insufficiently adapted to local conditions and research results do not come up with a variety of technological solutions adapted to the range of socio-economic and agro-ecological conditions existing in the country, such as the differing technical needs of female and male farmers. Lack of technological alternatives is often mentioned as a constraint to irrigation development (e.g. different models of irrigation pumps, suited to the needs of different users). Where techniques and technologies developed by research are available, their dissemination is faced with a number of difficulties such as the poor delivery of the extension and training services that are not necessarily targeted to the appropriate users.

69. Weak extension and training services and the consequent lack of technological knowledge of farmers are often considered to be the major factors behind the insufficient adoption of improved technologies. This constraint could be overcome by improving farmers' access to knowledge. For example, valuable information can be obtained from some of the extension materials on FAO's Ecoport web pages.

(v) Social and cultural factors

70. The development and adoption of high-production technology has also been constrained by a number of social and cultural factors, including:

• Insecurity of land tenure and fragmentation of land holdings in some LDCs, particularly in Africa, especially with regard to women, who may have little or no access to land, depending on custom or formal laws that regulate the tenure practices;
• The low level of education, which is an obstacle to raising the technological capacity of farmers from its currently low levels and to the adoption of new technologies. Education for both boys, girls and adult women is often lacking. This constraint is considered to be particularly acute for women. For example, the lack of farmers' bookkeeping skills makes it more difficult for them to appreciate the advantages of improved technologies;
• The risk-averse tendencies of farmers, which have been generally underestimated, particularly when they have not been involved in the decision-making process on the development and use of new products. Farmers have sometimes hampered the adoption of new technologies and management practices, especially when their traditional livelihoods and associated local traditions have been threatened. For example, high variability of yield of certain improved varieties has been a constraint to their adoption by poor farmers bordering on the subsistence level. It is essential that both male and female farmers be involved in the entire process of developing new high-yield varieties and associated technologies, in order to ensure a greater acceptance and adoption by those who stand to benefit most;
• Since educational attainment has a direct impact on the knowledge, skills, attitudes and behaviour of farmers, low educational attainment in LDCs is reflected in the great difficulty in changing farmers' attitude and behaviour to cope with the new policy reforms. The reforms implemented since the early 1980s have involved drastic changes in the environment in which producers operate. While they were relatively passive actors before policy reform, they have now to take initiatives and organize themselves.

2.3.3 Post-production activities

71. Lack of good quality roads as well as insufficient storage facilities have been identified as major constraints in many LDCs, sometimes resulting in crops remaining unsold. Failure of the transport infrastructure in some LDCs to move food grains from surplus to deficit areas during periods of localised drought illustrates the severity of transport bottlenecks and agricultural market segmentation. Inadequate communication facilities tend to limit, for many producers, the possibilities of access to markets and market information, as well as to make access to inputs more difficult and costly, and lowers producers' returns. The absence of storage facilities amplifies seasonal market fluctuations and the level of post-harvest losses, that in some cases can be as high as 30 percent of total production.

2.3.4 Food safety and quality standards

72. Ensuring the safety and quality of foods in developing countries is of paramount importance not only from the point of view of public health but also to improve the competitiveness of their food products in the international market. Their control systems and institutions suffer from a number of weaknesses which make them ineffective in ensuring consumer protection and benefiting from the post-Uruguay Round trading regime. These weaknesses concern all the basic elements of a national food control system, i.e. food legislation, food inspection, quality assurance at the production level and testing capabilities (human and physical) to control the quality and safety of the food supply.

73. The following actions are needed to enhance the capacity of developing countries to meet the requirements set out in the relevant WTO Agreements, thus ensuring consumer protection and promoting food trade, internally and externally:

• Capacity building to implement the Agreement on the Application of Sanitary and Phytosanitary Measures (SPS) and the Agreement on Technical Barriers to Trade (TBTs) and to fulfil other new international and domestic requirements on standards;
• Developing national capacities for risk assessment; designing domestic regulations and policies for export, food and agribusiness development;
• Promoting regional cooperation and enhancing coherence in trade policies and domestic regulations on standards;
• Development of national food safety regulations and standards without distorting international trade;
• Enhancing the participation of developing countries in international standard setting bodies;
• Collection of relevant information for national capacity building and policy making;
• Improving the negotiating stance in international trade negotiations

2.3.5 Investment in agriculture

74. Least developed countries face a major domestic resource gap in generating the investments needed to achieve their developmental objectives in agriculture, including the target of reducing the number of under-nourished people by 2015. The concept of investment to augment the productive capacity of agriculture entails not only physical assets, but also science and technology dissemination, human capital enhancement and social capital build-up. Creating a pro-investment climate to raise productivity levels and achieve the necessary structural changes is a major policy challenge.

75. In many LDCs, much public expenditure on agriculture is in the form of subsidies, leaving little public funding for the creation of new assets, for maintenance or for other growth-producing expenditure. The result is that many agricultural support services barely function, rural roads are impassable for much of the year, farm machinery is mostly inoperable and irrigation schemes are crippled.

76. It was seen above (paragraphs 53-59) that many LDCs have adopted policies to deregulate agricultural markets, reduce price distortions, and allow a greater role for private agents in economic activity. Such measures, although necessary, are not always sufficient for inducing the investment necessary to permit sustained production increases. Improved investment incentives also require policies that improve access to markets, ensure dissemination of information, set standards and provide an adequate legal and regulatory framework. At a more general level, there is consensus that political stability and a well-defined and enforced institutional framework are also needed. Strong complementarily between public and private investment is also necessary to sustain agricultural growth, with governments investing in sectors having an important public good element such as research, extension and infrastructure and hence covering in particular roads, education, norms and standards.

77. FAO projections put the average annual gross investment requirements until 2010 in SSA for primary agriculture, storage and processing (excluding the related pre- and post-production infrastructures and services) at some US$ 11.1 billion, under the "business-as-usual" scenario¹³. To reach the World Food Summit target of halving under-nutrition, however, additional investments of about US$ 3.6 billion would be required to ensure the necessary increase in domestic food production.

78. Rural infrastructure (e.g. irrigation and roads) is badly lacking in LDCs, particularly in SSA. Heavy investments are needed in rural communication infrastructure, irrigation improvements and modernization, better exploitation of rainfall by simple and improved water capture and use, land management and improvements, education extension and research and the provision of health services. Research results and best practices, as yet untried on a large scale, can be adopted for high-potential areas in SSA. Research continues to be needed to develop farming systems for small, resource-poor farmers, who form the majority in most LDCs. These new farming systems need to be sustainable at higher population densities and capable of intensifying production on existing land.

79. While the need for investment is huge, sequential removal of constraints is crucial. National experience has shown that when the increase in production associated with a reform or a change in circumstances has reached a plateau, a further reform, or series of reforms, has been required to unleash further potential. Sustained growth has only been possible when new constraints have been alleviated by further reforms.

80. Most of the required investments can be expected to be forthcoming from the private sector, and therefore depend decisively on the prevalence of a conducive climate at the national and international levels. In this regard, public investment is an indispensable pre-condition and catalyst for and complement to private investment, involving basically investment in research and infrastructure.

B. DEMAND ISSUES

1. Trends in consumption

81. Domestic consumption (human consumption and other uses) of agricultural products in LDCs varies widely between food and non-food products. Non-food products such as raw materials and tropical beverages are basically produced for export. The little that goes to the domestic market is destined essentially for local processing industries, which in turn export the bulk of their produce. In contrast, the domestic consumption of food products is a large and growing proportion of output. Consumption of basic foodstuffs in LDCs grew by an annual 2.3 percent during 1990-97 (Annex Table 1), below the population growth rate of 2.6 percent. The consumption of cereals met by domestic production declined from 96 percent in 1970-80 to 85 percent in 1990-98.

82. For many commodities, production has not, and perhaps will not, keep up with demand. For example, during the 1960s LDCs were net exporters of rice (2.4 million tonnes), but by the mid-1990s they were importing 3.5 million tonnes, a figure that is projected to rise to over 7.5 million tonnes by 2015. Similarly, net imports of wheat increased from 1.1 million tonnes in 1961-63 to 6.1 million tonnes in 1995-97 and are projected to reach 15 million tonnes by 2015. Cassava and plantains, the main staple food in many African LDCs, also showed an increase in net imports in the 1990s.

83. In sum, trends in production, consumption and trade amply demonstrate the increasing import dependence of LDCs for food. FAO projections for 2015 suggest that this dependence will continue to increase. If the requisite commercial imports cannot be ensured, or if food aid cannot make up for the shortfall, per caput food consumption will inevitably fall.

2. Determining factors and constraints

84. There are three determinants of demand growth: population, per capita income, and the income elasticity of demand. For LDCs as a whole, the real GNP per caput has been stable over the last two decades. Between 1989-91 and 1995-97, in only 20 out of the 40 LDCs for which comparable data are available did it rise. This suggests that the increase in cereal consumption in most LDCs during 1990-98 was due mainly to population growth. Population growth rates in LDCs are among the highest in the world. For LDCs as a whole, the rate increased from an annual 2.5 percent in 1980-90 to 2.6 percent in 1990-96 and is projected to remain high (2.3 percent) during 2000-2015. Their food security is thus threatened unless production performance and/or food import capacity can be improved significantly over the levels of the past 10 years.

85. The capacity to import food is determined by the availability of foreign exchange, which in turn is determined by export earnings (essentially from commodities for most LDCs) and by the external resource flow. Many LDCs suffered because of the fall in prices of their primary commodity exports during 1990-98. The foreign debt burden also limits the ability of many LDCs to import, and the situation has been compounded by the slowing down of the external resource flow.

3. Food security

86. The interaction between food supply and demand factors determines the level of food adequacy. The most widely available and used indicator for estimating food adequacy levels is per caput dietary energy supply (DES), which measures the food available to each person on average in a country. As shown in Tables 2 and Annex Table 8, the DES for LDCs as a group has been very low and barely risen since 1979. For roughly half of the 44 LDCs for which data are available it has been below 2100 kcal/day. This stands in contrast to the progress in other developing countries and the world as a whole, where food production has continued to outstrip population growth.

Table 2. Per caput dietary energy supply (DES)

* Excluding Bhutan, Equatorial Guinea, Samoa and Tuvalu, for which data were not available.

Source: FAO.

87. FAO estimates show that the incidence of chronic undernutrition (undernourishment)¹⁴ is high in LDCs (Annex Table 9). Between 1969-71 and 1996-98, the proportion of undernourished in total population in LDCs increased from 38 percent to 40 per cent, while the absolute number of undernourished is estimated to have increased from 116 million to 235 million.

C: INTERLINKAGES: AGRICULTURAL GROWTH, RURAL DEVELOPMENT AND POVERTY ALLEVIATION

1. NATURE OF THE LINKAGES

88. With 70 percent of the world's extremely poor and food-insecure people living in rural areas, the role of agricultural and rural development in the eradication of poverty and food insecurity is crucial. As agriculture is the predominant economic activity in rural areas, the rural poor strongly depend on it for their income and food entitlements. The dependence is most marked in countries where food insecurity is most widespread, and where there often exists a combination of low incomes, a food deficit and high external indebtedness. Most such countries are LDCs.

89. National experience of economic growth and poverty alleviation reveals that: i) poverty alleviation is positively related to overall economic development; ii) agricultural growth in developing countries has stronger effects on poverty alleviation than growth in other sectors; iii) it alleviates poverty mainly through the labour market, especially through increases in wages; iv) its impact on poverty reduction lessens if there is growing income inequality; and v) rural growth reduces both urban and rural poverty.

90. The potential for agricultural growth to alleviate rural poverty in the LDCs is exemplified by the fact that, on average, agriculture employs about 75 percent of the total labour force (over 80 percent in several cases) and that the percentage of poor in the rural areas is generally much higher than in the urban areas (see Annex Table 10). Hence, agricultural growth can increase the income of the poor both directly, through the additional demand for labour, and indirectly, through input, output and expenditure linkages with non-farm productive activities in the rural sector.

91. The rural non-farm sector constitutes the connecting link between agriculture, rural development and rural poverty alleviation. In many low-income countries, it is expected to be closely linked to agriculture in numerous upstream and downstream productive activities. Earnings from participation in such activities may constitute a substantial share of the overall income of rural populations. Annex Table 11 shows data on non-farm income and its distribution (when available) by (a) income percentiles (b) zones, according to the type of agricultural production; and (c) types of product cultivated. It covers a limited number of LDCs in Africa and Asia for which data are available.

92. Thus, farm and non-farm rural activities should be considered complementary in terms of financing investment in both sectors: savings derived from farm activities can constitute start-up capital for rural non-farm activities. At the same time, savings derived from non-farm activities can be used to acquire inputs and adopt improved agricultural technologies.

2. Does the type of agricultural growth matter?

93. The shares of non-farm income reported in the foregoing paragraphs demonstrate, if anything, that the terms rural and agricultural growth are not synonymous and that non-farm income is an important component in the livelihoods of rural households. Thus, when considering the impact of agricultural growth on poverty and rural development, its effects through production, income and expenditure linkages on rural non-farm income and employment should also be taken into account. Approaching the question of agricultural growth and poverty reduction within the more general rural development framework described above makes the analysis more complex.

94. One example is that of productivity-driven agricultural growth achieved through capital-intensive technologies. Such a pattern of productivity growth may not result in poverty alleviation for two reasons: (a) poor farmers lack the necessary access to capital that would enable them to benefit from the new technologies; and (b) agricultural growth is not translated into increased demand for labour and thus landless rural labourers do not benefit. Consequently, the agricultural growth does not directly benefit the poor. On the other hand, a more complete examination of the effects of agricultural growth should take into account the effects on rural incomes and poverty via the rural non-farm sector. Specifically, is increased agricultural output associated with increased demand for services provided at the local level (input provision or services or output processing and distribution)? Is additional income resulting from increased growth spent on locally produced goods?

95. In cases of extreme inequality in the distribution of productive assets and a capital-intensive technological change it can be expected that there will be no indirect effects (through various linkages of agriculture to the non-farm sector) and most probably the poor will not benefit. Agricultural inputs are likely to be "imported" (from urban areas or abroad) while the consumption patterns of those who benefit from agricultural expansion are likely to involve a large proportion of high-value commodities and luxuries that are not produced locally. Input, output and expenditure linkage effects can thus be expected to "leak out" of the rural areas.

96. On the other hand, the benefits of agricultural growth based on improvements in labour productivity are likely to be widely diffused in the rural areas. Such technologies (and the gains from them) may be accessible to poorer farmers, while landless labourers benefit from higher wages or employment. Input, output and expenditure linkages should favour the rural sector, since landless labourers and smallholders are likely to acquire inputs or services and spend additional income in the rural areas, thus increasing secondary income effects through the expansion of rural non-farm activity and demand for labour.

97. Another distinction concerning the types of agricultural growth concerns that of "food versus staples". The stylised fact that most of the rural poor derive income from the production of staples in the form of either food or other entitlements (i.e. income derived from employment in the production of staples or from activities linked to it) has prompted the "promotion" of staples production (in terms of research on ways of increasing staples yields) in preference to the production of cash or commercial crops. In the context discussed above, such an argument would imply that staples production has stronger linkages to the local economy and thus a stronger effect on reducing poverty and enhancing food security than non-staples. Such indeed may be the case for rural areas with limited access to food or other markets (such as urban and export markets). In such situations, linkages created by productivity-induced increases in food production are very strong, as there are no "leakage" effects.

98. Nevertheless, no general statement can be made in favour of or against staple commodities. There is no evidence that shifts to cash (or commercial) crops have been associated with increases in poverty. Cash crops have much to offer in the way of both higher income and greater income diversification opportunities. In the presence of higher risks associated with reliance on the market for both food and income, farm households can be expected to diversify their resources between them. Critical requirements for successful diversification opportunities are that channels for the supply of inputs and marketing of outputs are opened up and that there should exist well-functioning rural financial markets.

99. In sum, the role of agricultural development in overall economic development and in eradicating poverty and food insecurity in LDCs is crucial. Measures to that end include: raising agricultural productivity and encouraging other sources of rural development, notably through rural infrastructure; enhancing human capabilities in rural areas through health, education and sanitation services and access to productive resources, with stress on gender equality; and preserving the capacity of the natural environment to sustain the present population and future generations.

1 For example, recent World Bank figures for Sub-Saharan Africa (SSA), which contains the majority of LDCs (34), show that the proportion of poor people increased from 38.5 percent in the late 1980s to 39.1 percent in the mid-1990s. (World Bank, Entering the 21^st Century: World Development Report 1999/2000, (New York: Oxford University Press for the World Bank, 2000), p. 25).

2 See FAO (2000), Agriculture: Towards 2015/30, Technical Interim Report.

3 One of the clearest consequences of population increase will be to exert pressure for more land to be brought under cultivation. At the same time, existing agricultural land will be used more intensively.

4 Potential arable land as referred to here is a rough indicator: it includes lands which are currently under forest and wetlands which are protected and not available for agriculture and makes no allowance for land for human settlement. Thus, land potential as shown in Annex Table 2 is likely to be overestimated, but it should nevertheless be a good indicator of the relative potential of different countries.

5 Although yield comparisons should be in a homogenous agroecological context, such comparisons of averages provide a good idea of the range of possibilities.

6 IFPRI, "Is there hope for food plenty in Africa?", News & Views: A 2020 Vision for Food and the Environment, October 1996.

7 FAO, Rome (1995), The National Agricultural Research Systems of West and Central Africa.

8 See UNDP, Human Development Report 2000 (New York : Oxford University Press for UNDP), 2000.

9 Collomb, P. (1999): Une voie étroite pour la sécurité alimentaire d'ici à 2050. FAO, Rome, and Economica, Paris.

10 D. Topouzis and J. du Guerny, Sustainable agricultural/rural development and vulnerability to the AIDS epidemic, FAO/UNAIDS joint publication, 1999.

11 Dorward A. and Morrison J. (2000), "The Agricultural Development Experience of the Past 30 Years: Lessons for LDCs", background paper prepared for FAO.

12 Mellor J. (2000), "Agricultural Development: So many Successes, Such Excellent Results", background paper prepared for FAO.

13 FAO Committee on World Food Security, Investment in agriculture for food security: Situation and resource requirements to reach the World Food Summit Objectives, CFS: 99/Inf.7, June 1999.

14 The term "undernourished" in the context of the World Food Summit 1996 refers to persons whose food consumption level is inadequate in terms of calories consumed relative to requirements on a continuing basis.