Jump to ContentJump to Main Navigation
The Political Economy of Hunger: Volume 3: Endemic Hunger$

Jean Drèze and Amartya Sen

Print publication date: 1991

Print ISBN-13: 9780198286370

Published to Oxford Scholarship Online: January 2008

DOI: 10.1093/acprof:oso/9780198286370.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (oxford.universitypressscholarship.com). (c) Copyright Oxford University Press, 2020. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 11 August 2020

8 The Contribution of Industry to Solving the Food Problem in Africa

8 The Contribution of Industry to Solving the Food Problem in Africa

(p.281) 8 The Contribution of Industry to Solving the Food Problem in Africa*
The Political Economy of Hunger: Volume 3: Endemic Hunger

Wangwe Samuel

Oxford University Press

Abstract and Keywords

This chapter investigates some important intersectoral relationships, specifically the positive contribution of industry, by raising agricultural productivity and creating employment opportunities and hence increasing incomes outside agriculture, in alleviating the food problem in sub-Saharan Africa. It emphasizes the creation of linkages between agriculture and other sectors, linkages between industry and food production activities in agriculture, and the link between research and development. It also explores the role of industry in agricultural processing. To solve the persistent endemic undernourishment, the chapter suggests a wide range of remedial actions involving institutional changes and economic reforms, both within and outside the food sector.

Keywords:   food crisis, determinants of entitlement, industry, agricultural equipment, employment opportunities

8.1. Introduction

Recent studies on the economic conditions in Africa have portrayed a deepening crisis. An important aspect of this crisis is the food problem, visible not only in recurrent famines but also in high and possibly increasing levels of endemic undernutrition. Addressing the interrelated problems of poverty and hunger in an effective manner is one of the most urgent policy challenges facing African countries today.

In 1980 the share of the labour force engaged in agriculture in sub‐Saharan Africa was estimated at 72 per cent. For the 23 low‐income countries of sub‐Saharan Africa this share was even higher, averaging 78 per cent (World Bank 1984). In analysing the problem of hunger and poverty, the focus must be on the agricultural sector, but not in isolation from sectors related to it (e.g. industry). While rightly focusing on the problems of the agricultural sector, many recent studies of the food crisis in Africa have tended to analyse the problems of this sector in isolation.

A closer examination of agriculture and the food problem in Africa reveals that intersectoral relationships are pervasive and important. The purpose of this chapter is to investigate some of these interrelationships, with a view to bringing out the positive contribution of industry to alleviating the problem of hunger and poverty. The contribution of industry is basically twofold. First, industry plays an important role in raising productivity in agriculture. Second, it enhances access to food by creating incomes outside agriculture.

In analysing the problem of hunger and poverty, some scholars have focused on aggregate food supply, taking its distribution as given. Other scholars, led by Sen (1981), have focused on the issue of entitlement, suggesting that changes in the conditions of acquisition of food can have a considerable effect on the extent of hunger even for a given level of aggregate food supply. The issues of food supply and food access are obviously not unrelated. Given a particular distributional pattern, a decline in food supply per capita would (p.282) result in reduced entitlements for vulnerable groups and exacerbate hunger. Ultimately, entitlement analysis must pay due attention both to the mechanism through which the aggregate supply of food is distributed, and to the determination of aggregate supply itself. This chapter examines the positive contribution of industry to enhancing the entitlements of vulnerable groups, keeping in mind both the aggregative and the distributional elements of the determination of entitlements.

The plan of the chapter is as follows. Section 8.2 reviews some recent trends of the food situation in Africa. Section 8.3 discusses the major determinants of entitlements in sub‐Saharan Africa, and the role that industry can play in protecting and promoting them. Section 8.4 is devoted to a case‐study of agricultural equipment as a source of intersectoral linkages. Section 8.5 discusses the role of industry in agricultural processing. Section 8.6 offers some concluding comments.

8.2. Aggregate trends in the food situation

The deepening food crisis in Africa is partly evident in terms of aggregate indices from the declining growth rates of food production, declining food output per capita, increasing growth rates of food imports and food aid, and declining food self‐sufficiency ratios. These trends are shown briefly in this section.1

(a) Growth rates of food production

While food output increased at an annual rate of 2.5 per cent during 1960–70, it increased by only 1.7 per cent per annum during 1970–82 (World Bank 1984). As shown in Table 8.1, Africa performed below the world aggregate trends after 1970.

Table 8.1 Growth in agricultural production 1961–1980 (% p.a.)

1961–5 to 1970


1961–5 to 1980









Source: FAO (1981: 5).

The growth of population (about 3 per cent a year) considerably outstripped the growth of food output during 1970–82. Thus annual growth of food output per capita declined from 0.2 per cent during 1960–70 to only −0.9 per cent during 1970–82 (World Bank 1984).

The index of food production per capita in Africa (1969–71 = 100) was 93.9 in 1975, falling further to 88.3 in 1980 and 83.1 in 1983 (World Bank Tables). The index was above 100 in only 6 countries in 1980–2 (World Bank 1984).

These trends are alarming, not only because food production is an important source of food supply in Africa, but also because the production of food is a crucial source of income for much of its rural population.

(b) Food self‐sufficiency

During the 1970s, 39 out of 44 countries had lost in food self‐sufficiency (Burki 1985), while the food self‐sufficiency ratio fell from 98 per cent in the 1960s to (p.283) 86 per cent in 1980 (Burki 1985 citing World Food Council 1984). According to FAO (1981), in Africa self‐sufficiency ratios in cereals have fallen from 95 per cent in the early 1960s to 83 per cent in the late 1970s. By the year 2000 Africa's self‐sufficiency ratio (on the basis of present trends) could drop to 56 per cent.2 In FAO's view an annual food production growth rate of at least 4.3 per cent in the 1980–2000 period is required to reverse these declining trends.

During 1974–83 food imports (in cereals) increased at the rate of 9 per cent a year while food aid (in cereals) increased at 13.3 per cent per year. Food aid (in metric tons of grain equivalent) increased from 1,237,200 tons in 1978 to 2,168,600 tons in 1982 (i.e. about 5 per cent of total cereals production) (World Bank 1984). The aggregate food inflow (in cereals) from outside the region increased at 9.6 per cent a year during this period. When this trend is juxtaposed with worsening balance‐of‐payments problems and the increasing reluctance of donors to provide food aid, it would appear that entitlement to food at the national level (i.e. the ability of the nation as an entity to establish command over an adequate amount of food) is at stake.

Even when food imports and/or food aid are available, the logistics of distributing food to the hunger‐stricken areas have often been problematic, as demonstrated by the recent experiences of Sudan and Ethiopia (Prattley 1986, several African Emergency Reports). This suggests that a reasonable level of self‐sufficiency, and related food security arrangements, are necessary, especially in areas which are difficult to reach with food coming through ports. Food imports and food aid cannot be a reliable way of preventing loss of entitlements to food.

This poses at least two challenges. First, the need to increase domestic food production. Second, the need to revamp the export sector so that the expanded export revenues can be used to import food or to improve production and processing technology in the food‐producing sectors.

The options here will reflect country‐specific situations and the pertaining long‐term comparative advantages. On the food production side, care should be taken against trying to achieve food self‐sufficiency at any cost, e.g. by growing crops which are not appropriate for climatic conditions in Africa. For example, Andrae and Beckman (1985) have indicated that in Nigeria the (p.284) illusion of food import substitution has sustained a massive waste of national resources on large‐scale, technologically advanced, import intensive wheat production which can do little to solve Nigeria's food crisis and dependence on imports. On the side of exports, the predominance of traditional exports and their deteriorating terms of trade point to the need to examine possibilities of developing non‐traditional exports if export revenues have to increase substantially in future. Decisions as to what exports to promote can only be country specific but the principle of avoiding expansion of the output of traditional exports which are characterized by a low income elasticity of demand seems to be worth adopting in the formulation of export strategies in Africa.

8.3. Challenges of the entitlement problem

This section discusses the major determinants of entitlements in sub‐Saharan Africa, and the possible causes of loss of entitlements. An attempt is also made to bring out the positive contribution of industry to solving the hunger problem, both by raising productivity in agriculture and by enhancing the entitlements of those who derive their incomes from non‐agricultural activities. Considering the limitations of data on sub‐Saharan Africa as a whole, this study will be based on a sample of case‐studies from selected countries in sub‐Saharan Africa.

(a) Loss of harvest

Those groups in society who acquire food by producing it (either as owner‐cultivators or as cultivating tenants) usually lose their command over food in the event of loss of harvest. For these groups, preventing crop failures is important for safeguarding their command over food. In this context, it has been pointed out that various risk‐reducing practices, such as mixed cropping, combination of upland and valley plots, or cultivation of drought‐resistant crops, can contribute to the security of farmers in spite of reducing average yields because they reduce the variation of total farm output (Berry 1984, Richards 1983, Norman et al. 1979). These practices reduce the risk of entitlement failures for farmers. Industry can further enhance the entitlements of this group by generating employment and incomes in non‐agricultural activities, or through food preservation and processing of foodstuffs. These issues are taken up in the appropriate sections of this paper.

(b) Marketable surplus

Food‐producing groups often produce a surplus over and above their food requirements. If this happens, then food can be made available to the other groups. As the size of groups which rely on purchasing food increases (e.g. due to urbanization), the need to increase marketable surplus from the food‐producing groups increases. The response to this demand should take the form of expanding production while at the same time enhancing or at least safe‐ (p.285) guarding the consumption levels of farmers. The other alternative of obtaining food through imports is feasible only if a viable export sector can be developed.

Marketable surplus can partly be stimulated by increased supply of relevant industrial goods for peasants to purchase. There are at least two important categories of such goods. First, the supply of farm inputs from the industrial sector can play the role of enhancing the capacity of the food‐producing groups to produce more food (e.g. by employing more productive agricultural inputs and implements produced in the industrial sector). In section 8.4 it is shown that much of industry in Africa has not been geared to supply agriculture with the equipment and inputs it requires. Second, the supply of ‘incentive goods’, especially manufactured consumer goods, can enhance the willingness of the food‐producing groups to produce food for sale in the market. For instance, studies on Tanzania and Mozambique have revealed that the non‐availability of incentive goods in rural areas has had a negative effect on the size of marketed output. Such experiences suggest that the availability of incentive goods (mainly industrial) is an important component of the policy package for increasing marketable surplus.

The role of food prices in this policy package has been controversial. If food prices are kept low, efforts to increase food output may be reduced but the ability of vulnerable groups to acquire food may be enhanced. If prices are kept high, food producers' efforts may be stepped up but the poor groups in society may not have access to the more expensive food. The solution of subsidizing food often runs out of steam in the face of budget deficits. However, Mellor (1986) has argued that rising food prices are usually indicators of supply problems arising from sluggish technological change, while declining food prices indicate success in inducing rapid technological change. This suggests that helping food producers to increase output and marketable surplus requires not only price incentives but also agricultural inputs, innovations (e.g. new seeds, new technology), information (e.g. extension services), infrastructure (e.g. roads, irrigation), and effective institutions (e.g. marketing, credit, land reform). Industry can make an important contribution to increasing marketable surplus at reasonable prices by promoting the availability of some of these factors.

(c) Allocation of investment resources

The share of agriculture in development resources has been low in Africa, and within the agricultural sector the resources have been allocated in such a way that small farmers have hardly benefited from them.

On the allocation of resources to agriculture, Lipton (1985) citing FAO (1983) and UN (1983) has presented some data on the share of gross fixed investment to agriculture in 10 sub‐Saharan countries. The results, given in Table 8.2, indicate that this share was below 10 per cent in seven countries. In four countries the share of investment allocated to agriculture was not more than 3 per cent.

Table 8.2 The share of gross fixed investment allocated to agriculture


% of gross fixed investment allocated to agriculture

% of the workforce engaged in agriculture































Source: Lipton (1985).


Within the agricultural sector, governments have tended to allocate resources so as to support the production of marketable agricultural surplus. The question is whether and to what extent this policy has assisted small farmers in consolidating their entitlements to food.

This question is approached here by considering the agricultural sector in terms of two subsectors. One is the traditional sector (hereafter sector T), dominated by small‐scale farming peasants employing a low level of technology (usually based on hand‐tools and to some extent on animal‐drawn equipment). The other is the modern farming sector (hereafter sector M), dominated by large‐scale private or state farms employing a high level of technology (usually based on imported or high import content motorized machinery) and providing the bulk of the agricultural sector wage employment. Fig. 8.1 depicts the two subsectors.

8 The Contribution of Industry to Solving the Food Problem in Africa

Fig. 8.1. Production and productivity in two subsectors

Note: Lt: total labour force in agriculture; sector M: modern farming sector; sector T: traditional farming sector.

As capital accumulation takes place in sector M, line AC shifts to the right, but OL t also shifts to the right towards L t+i. Since the proportion of the labour force in sector T is 70–80 per cent of the total labour force in Africa, ‘dualism’ (in the sense of a marked difference in labour productivity in the two sectors) is likely to be perpetuated. In economies where capital accumulation in sector M is accompanied by a fast expansion of employment in other sectors of the economy (e.g. industry, services), the absorption of labour in these sectors may be rapid enough to prevent a shift of OL t to the right towards L t+i. L t may even shift to the left. This is what happened in the now developed countries in Europe and North America. In the context of Africa, however, the relative size of CL t is high (70–80 per cent of the total labour force and about 98 per cent of the labour force in agriculture), and the rate at which non‐agricultural sectors are absorbing labour is very low.


Under these conditions, investment in sector T to upgrade the level of technology and raise labour productivity is likely to be a more appropriate strategy than further investment in sector M for two reasons. First, the degree to which investments in sector M are increasing employment is far below the degree to which the population in sector T is increasing. This partly explains the high rate of urbanization which has outpaced even the growth of employment in the non‐agricultural sectors. Further investments in sector M would increase agricultural output and labour productivity in this sector but would not increase output, productivity, and incomes in sector T where 70–80 per cent of the labour force is employed. In other words, that kind of investment decision would not necessarily enhance entitlements to food for 70–80 per cent of the population. In fact, in some cases investments in sector M have resulted in loss of entitlement on the part of some groups in sector T. For instance, in the case of Nigeria, Andrae and Beckman (1985) have shown that, as a result of the Bakolori project (sector M), some groups (in sector T) lost their land, were prevented from cultivating their farms for two or three consecutive seasons, and in some cases even had their crops destroyed by contractors. Second, investments in sector T are likely to affect the majority of the low‐income earners more directly by increasing their productivity.

In practice, however, the allocation of resources in Africa has usually favoured sector M. A number of studies have shown how the modern sector bias of resource allocation by governments, while creating entitlements for some wage‐earning groups in agriculture, has led to the neglect of large groups of small‐scale farmers. It has also been suggested that post‐colonial governments have demonstrated a common inclination to allocate resources to state (p.288) farms, parastatal enterprises, and joint government–private ventures in large‐scale farming, often providing generous loans, subsidies, infrastructure, and technical assistance to a small number of large modern farms (Berry 1984; Heyer et al. 1981; Hill 1977; Samoff 1980; Bates 1981). For instance, it has been estimated that of the little public expenditure allocated to agriculture (often less than 10 per cent of the total) about 90 per cent has been allocated to the modern farming sector. In the area of research, very little has been directed to the food crops which are usually grown by the majority of small farmers. For instance, Eicher (1986) has cited Spencer (1986) on the point that less than 2 per cent of total sorghum, millet, and upland rice area in West Africa is sown to cultivars developed through modern genetic research. Although maize is more promising in East and Southern Africa, improved seeds have not been adequately supplied to smallholder farms. In the case of fertilizer consumption, IFPRI (1986) has indicated that although smallholder farms account for 75 per cent of Kenya's total agricultural output they consume only 42 per cent of total fertilizer supply. Similarly, about 80 per cent of agricultural equipment imports are destined for the modern farming sector (state farms, plantations, and agro‐food complexes). The majority of peasants (70–80 per cent of population) receive direct benefit from less than 5 per cent of agricultural equipment imported (UNIDO 1983 a: 81).

Moreover, the efficiency with which the resources allocated to sector M are used is not always impressive. The levels of utilization of modern equipment has been low, even though its importation consumes scarce foreign exchange. For instance, according to UNIDO (1983 a), a report from the Kenyan Ministry of Agriculture shows that the level of productive utilization of tractors from Kenya's Tractor Hire Service was only 7 per cent. Similarly low rates have been recorded in many countries of sub‐Saharan Africa.

Attempts to combine high‐level mechanization with the peasant economy have not been very successful either. The contradictions pointed out by Andrae and Beckman (1985) in the case of wheat production in Nigeria, between the social organization implicit in a high level of mechanization and that of the peasant economy, are quite instructive. For instance, they have pointed out that mobile combine harvesters were turned to stationary combine harvesters which had to be fed manually by farmers who cut their wheat with sickles. The high‐capacity combines are made to perform the work of a simple stationary thresher partly because the farmers' plots are too small to be harvested by a combine if each peasant is to receive his own crop.

In this section it has been shown that only a very small proportion of total investment is channelled into agriculture. Out of this small investment only a small proportion is channelled into the smallholder farming community in spite of this community's high share of the total labour force. In order to enhance the productivity of smallholder farmers and enhance their entitlement to food there is a strong case for channelling more resources into the smallholder farm sector. In the light of interaction of household decisions as both (p.289) producer and consumer as demonstrated, for instance, by Singh et al. (1986) the numerous linkages between agriculture and other sectors (e.g. through investments, consumption, production, foreign trade, and employment) must be taken into consideration. Support towards raising productivity and incomes of the smallholder farmers will consist of a broader policy package than direct investments in agriculture. In this context investments in rural industry, in rural workshops for manufacturing and repairing farm tools and equipment, rural‐based agricultural processing activities, rural transport systems, and investment in employment intensive and local resource‐using activities producing incentive goods would result in the growth of productivity, incomes, and employment in the bulk of the rural areas in Africa.

(d) Wage employment in agriculture

Agricultural wage earners are usually employed on large farms (private or public). For some of them the wage earned is all the entitlement they have to exchange for food while for others this is supplementary income. This group is a product of the process of agrarian change over time in many ways. In some cases rural poverty compelled people to seek off‐farm employment (Hill 1972). Such rural poverty may be due to loss of access to an adequate amount of land (or to any at all) or to the loss of harvest due to drought as in the case of parts of Ethiopia (Burki 1985; Sen 1981). In many cases, however, this group consists of members of the family who migrate to work on the large farms and leave behind the rest of their family members on the small farm. This has been a way of ensuring that resources generated from the land are supplemented by wage income in a way which reduces risk of loss of entitlement in case of crop failure. In some cases this additional income has been used to help finance purchases of rural land or to increase the peasant production for the market (Collier and Lal 1980; Berry 1984).

There are, thus, diverse reasons for the expansion of wage employment in African agriculture. While some of these reasons relate to the impoverishment of some sections of the rural population, it is also important to recognize the positive contribution of wage employment to the diversification of rural incomes. The development of non‐agricultural or off‐farm employment in rural areas, e.g. by relocation of some industries in these areas, can play a similarly positive role.

(e) Off‐farm activities

In order to safeguard their entitlements in the event of crop failure, some peasants engage in off‐farm economic activities for extra income. Other groups who were formerly peasants and lost their entitlement to land have no choice but to seek alternative employment. Since wage employment opportunities in other sectors are limited, these groups are often forced to turn to non‐farm activities to earn a living.

As regards the first group (i.e. peasants who seek to supplement their farm (p.290) incomes), their entitlements to food have tended to be enhanced by their non‐farm activities. For instance, a study of smallholding households in central Kenya has shown that on average nearly half of their annual cash incomes were derived from off‐farm sources (Njonjo 1981). In fact, one of the main observations made in the predominantly smallholder zone in Kenya is the growth of private services, a trend which brings out the linkages between agricultural production and non‐agricultural employment (IFPRI 1986; Heyer in Chapter 7 above). Berry (1984) has pointed out that income from off‐farm employment is a kind of windfall gain for rural households whose members seek a vent for their surplus labour in the dry season by migrating (Prothero 1975). For instance, Byerlee et al. (1983) found that adults in rural areas in Sierra Leone worked an average of 70 hours per month in the dry season, mainly on small‐scale manufacturing which was for them an important source of dry‐ season employment.

The tendency to diversify sources of income has been noted even among the surplus earning farmers who tend to diversify their portfolios using the proceeds of their farms to invest in trade and urban real estate rather than in expanded agricultural production (Berry 1984; Boesen and Mohele 1979; Cliffe 1978). This may be a reflection of greater profit‐making opportunities in the tertiary sector, as suggested by Berry (1984), but it may also be an expression of the importance of spreading the sources of income for purposes of reducing the risk of loss of entitlements. Thus, there are also important linkages between agriculture and other sectors through the processes of saving and investment. The tendency of surplus‐earning farmers to diversify their portfolios into non‐agricultural activities suggests that, if conditions were favourable, these investments could be channelled into the industrial activities which are complementary to agriculture. But as shown in sections 8.4 and 8.5 this has largely not been the case in practice.

(f) Employment in industry

In the African context, industrial employment has remained quite small. Industrial sector employment constitutes only 3–5 per cent of the labour force in sub‐Saharan Africa. Moreover, the rate of growth of industrial employment has lagged behind that of industrial output. The increasing use of capital intensive techniques of production is one of the reasons why the contribution of industry to enhancing entitlements through employment has been limited.

A more important observation in our context is the nature of the relationship between the development of large‐scale industry on the one hand, and rural small‐scale and cottage industries on the other. The relationship between them has been more competitive than complementary, with large‐scale industry developing at the expense of, rather than in support of, rural industries. In Tanzania, for instance, Muller (1978) has indicated that rural‐based metal‐working skills degenerated when large‐scale farm implements came into use, while in Kenya, Kongstad (1980) has pointed out that mechanical workshops (p.291) in non‐urban areas are degenerating partly because of the influx of products from the large Nairobi‐based factory. In the case of Ethiopia, Mohammed (1980) has referred to the same phenomenon in the textile sector. This particular tendency has resulted in a loss of incomes for those who were engaged in such small‐scale industrial activities. To the extent that individuals in the food production sector were partially engaged in these small industrial activities, this loss makes them more vulnerable to loss of entitlements in the event of crop failure and/or depresses their purchasing power (e.g. for agricultural inputs and consumer goods).

Two policy implications arise from these observations. First, there is a need to make rural industries and large‐scale industries complementary, e.g. through contracting. Second, the linkage between industry and food production activities in agriculture needs to be developed.

8.4. Agricultural equipment and inputs

The agricultural sectors of most African countries have experienced shortages of agricultural equipment and inputs. One report (World Bank 1981) has suggested that this shortage is caused by weaknesses in distribution. It has argued that the distribution of these equipments and inputs has been in the hands of state organizations and that these state organizations have not carried out this function effectively and efficiently. The problem of shortage of agricultural equipment and inputs in Africa, however, goes beyond the sphere of distribution. The experience of most African countries suggests that in analysing this problem the otherwise well‐known relationship between industry and agriculture has been neglected. This has largely inhibited industry from contributing to the alleviation of hunger and poverty through increased productivity in agriculture and in food production in particular.

In order to bring out this phenomenon, this section will examine one case‐study, namely production of agricultural equipment. The choice of study of agricultural equipment is not necessarily a reflection of its importance relative to other agricultural inputs. The findings on the manufacture of agricultural equipment will provide useful lessons on the role of industry in developing agricultural technology, a role with three distinct benefits: first, increasing productivity in agriculture, thereby increasing the supply of food among other things; second, increasing the productivity of the low‐income small farmers; third, creating incomes for those who are engaged in the industries which manufacture and repair agricultural equipment. The findings from this case‐study are largely applicable for other types of agricultural inputs too.

(a) Level and structure of production

The experience of most countries in Africa suggests that the inadequate supply of equipment to agriculture reflects sectoral imbalances. The industrial (p.292) sectors, even in countries where industry is relatively well developed (e.g. Nigeria) have not been linked to the agricultural sectors in terms of supplying the latter with equipment. The domestic industrial sector produces a very small share of all equipment actually used in agriculture. Africa's own agricultural machinery sector employs less than 1 per cent of the total industrial labour force. Its output accounts for less than 10 per cent of the total market for agricultural machinery (UNIDO 1983 a). In fact, the study of UNIDO (1983 a)3 suggests that the present indigenous agricultural machinery industry in most African countries is not only small but also in such poor shape financially and technologically that its very survival is in doubt.

As far as the problems faced by manufacturers of agricultural equipment in Africa are concerned, experience has shown that:

  1. 1. even though a demand often exists, local producers take only a small share of it and have so far been unable to contribute anything more;

  2. 2. they lack the ability to compete with imports and they are often not protected from these imports,

  3. 3. they lack the knowledge to supply the market with what it really needs.

As a consequence, some firms underutilize their capacity while others diversify into other lines of business.

Links between the existing basic industrial installations and agricultural machinery companies are very limited. This is in spite of the fact that nearly all the industrial units studied by UNIDO (1983 a) in the sector were set up during the 1950–72 period. They benefit from long experience but in many cases they operate almost in isolation of other engineering activities in the local industrial sector.

In recent years only a few units have been established, while many have either disappeared, or been merged with others, or have abandoned production of agricultural equipment in their manufacturing programme. Only about one‐third of the companies exclusively produce agricultural equipment. The others have diversified production into other product lines or are producers coming from outside the sector.

Via their subsidiaries and agents it is the Trans National Corporations that supply most of the imported equipment. TNCs' local production takes place in response to government pressure and is usually limited to final assembly of imported pre‐assembled units. Because of the importance of the size of the market these are found in the larger African countries. Their emphasis in local production is always on tractors and heavy equipment for land improvement. (p.293) The local producers' share (the rate of self‐sufficiency) is thus 15 per cent based on final value and only 5 per cent based on added value (UNIDO 1983 a).

The level of development of the agricultural equipment manufacturing sector is low both in relation to total industrial output and in relation to total demand. The structure of the sector is biased in favour of assembly of equipment for sector M with weak, if any, links with the rest of the domestic industrial sector.

(b) Imports of agricultural equipment

To the extent that the domestic industrial sector is not geared to the production of agricultural equipment, most of the demand for this has had to be met by imports. Thus imported agricultural equipment constitutes about 90 per cent of total supplies. As shown in Table 8.3, tractors constitute 60–70 per cent of all imports of agricultural machinery. In world terms, Africa shows a greater than average emphasis on tractors.4 Import data show a rapid increase in agricultural machinery up to 1975 followed by a series of erratic moves in the second half of the 1970s.

Table 8.3 Africa's imports of agricultural machinery, 1973–1979 ($000)



Hand‐ toolsa





All developing Africad of which:





Sub‐Saharan Africa











All developing Africad of which:





Sub‐Saharan Africa











All developing Africad of which:





Sub‐Saharan Africa











All developing Africad of which:





Sub‐Saharan Africa











All developing Africad of which:





Sub‐Saharan Africa











All developing Africad of which:





Sub‐Saharan Africa











All developing Africad of which:





Sub‐Saharan Africa










(a) Standard International Trade Classification 695.1.

(b) Standard International Trade Classification 712.5.

(c) Standard International Trade Classification 712 less 712.5.

(d) Excludes only South Africa.

Note: LDCs = Least developed countries in sub‐Saharan Africa.

Source: United Nations Statistical Office, New York.

Another characteristic of imports of agricultural equipment is the proliferation of the importing channels in most countries in Africa. There is a tendency to allow too many import companies to operate, resulting in numerous models irrespective of the size of the market. For instance, UNIDO (1983 a) has reported that in Sudan there are 20 companies importing tractors for a market of 575 units (1980), while in Zambia there are 17 companies from 15 different countries importing a total of less than 800 units. In many countries in Africa this tendency exacerbates the problems of maintenance and supplies of spare parts. The situation is further aggravated by the weak link between the activities of these importing channels and the local engineering activities. These factors play an important part in the observed low levels of tractor utilization.

These observations point to the need to formulate import policies which (1) encourage selective agricultural equipment imports with a view to supporting local industry and maintenance (possibly by stimulating demand through local manufacture of parts and components as well as maintenance and repair workshops), and (2) protect infant industries which are engaged in the production of various types of agricultural equipment.

(c) The link with research and development

It has been observed that most local manufactures of agricultural equipment in Africa have been assemblies of imported components. To the extent that this is the case, these manufactures need not coincide with the real needs of the agricultural sector in Africa. One way of linking local manufacture of agricultural equipment to the needs of farmers in sector M, and even more (p.294) importantly sector T, is through local research and development. A strong link between local research and development units and the farmers on the one hand and the local manufacturers of this equipment on the other is necessary. In this context UNIDO (1983 a) examined the status of R. & D. activities and (p.295) observed that there is little R. & D. capability at the producer level, which constitutes a major barrier to technological progress and prevents design and adaptation of machinery to local needs. Furthermore, it has been noted that there are no subregional or regional organizations (or associations of national organizations) devoted to designing, testing, and manufacturing agricultural machinery (whereas in, say, Southern Asia there is a Regional Network for Agricultural Machinery).

Most of the activities of the agriculture‐oriented research and development centres principally concern testing of imported equipment. Efforts are often duplicated (intra‐ and inter‐country), there is little exchange of prototypes, little co‐ordination of programmes, and most prototypes remain as prototypes (because of the weak link with manufacturing or lack of financing). Weak R. & D. facilities combined with a weak base of local production of agricultural equipment make it very difficult to respond to the needs of the majority of small farmers in sector T.

The results of the interaction between weak R. & D. and a weak manufacturing base have been demonstrated in a number of studies. For instance, the report of the Commonwealth Secretariat (1979 a) found that even where locally developed implements were being tested the absence of local workshop/manufacturing capacity inhibited production of the recommended modifications and adaptations. On the limited manufacturing capacity, the workshop5 noted inter alia that although the technique of manufacturing mouldboard ploughs is not sophisticated there are few workshops where these are being assembled or manufactured in Africa. Kaul (1979) has further indicated that, according to research carried out by the Institute of Agricultural Research at Zaria, implements introduced to Nigeria (e.g. Ariana, Unibar) have a potential use but have not been widely adopted due to lack of availability and inadequate users' training. An earlier Report of the Rural Technology Meet (Commonwealth Secretariat 1977) indicated that in response to the growing realization of the need for low‐cost, locally made, and easily managed implements, R. & D. institutions in many countries of Africa have developed new models of animal‐drawn ploughs and planters, water pumps, windmills, carts, and other types of simple equipment. However, the report goes further to indicate that with few exceptions the designs have not been manufactured on a large scale and have not yet been widely accepted by farmers. This is in spite of the benefits of using this equipment. In his study of the economics of animal traction in Burkina Faso, Jaeger (1985) observed acreage increases of 30 per cent where the weeding implements were used, and yield increases were found where the ox‐plough was dominant. A further study by Kjaerby (1983) found that in Tanzania users of ploughs have higher earnings than users of hoes, while in addition he cites studies from other countries which show that the plough made the work load for women relatively lighter in The Gambia (p.296) (Mettrick 1978) and Senegal (Venema 1980). These studies demonstrate that the adoption of appropriate agricultural equipment is capable of increasing the earnings of the small farmer. What seems to be lacking, as was mentioned earlier, is a strong link between R. & D. and the farmers (especially the small farmers) on the one hand and the local manufacturing units on the other.

In examining the process involved between perceiving specific needs of target groups in society and actually meeting these needs through local production of agricultural equipment, the six stages suggested by Wright (1977) are useful in locating the obstacles in this process. These stages are as follows:

  1. 1. analysis of needs, resources, and constraints;

  2. 2. search for suitable existing equipment;

  3. 3. design, construction, and testing of experimental prototypes;

  4. 4. field testing of production prototypes;

  5. 5. limited adoption;

  6. 6. widespread adoption.

While numerous programmes in Africa have reached stages (3) or (4), few have gone beyond. Wright examined those cases which had failed to go beyond stages (3) or (4) and those few which had gone through all the stages with a view to identifying reasons for success or failure in this context. He found that cases which have not gone beyond stages (3) or (4) were characterized by failure to prove their economic viability in the hands of local farmers, or were inhibited by limited local manufacturing facilities. The successful cases were both economically viable and enjoyed the support of the available local manufacturing facilities.

The co‐ordination of inputs, Wright (1977) observes, has proved to be another stumbling block in moving through the six stages. Citing the experience of Tanzania, Wright indicates that while testing and development of agricultural equipment was carried out by TAMTU (Tanzania Agricultural Machinery Testing Unit), nobody seemed to be willing and able to manufacture the items the unit had successfully developed. The large factory (UFI) was already utilizing all its capacity to meet demands for standard hand‐tools and the mould plough. TAMTU therefore set up a limited number of pilot rural craft workshops in several regions of the country. These, however, being run as government departments could not be operated commercially and often ran into financial difficulties. Beeney (1975) has evaluated the performance of these rural craft workshops and has observed that they suffered from lack of financial support and positive direction. Some had too little physical space (100 square metres) and their machine tools were too small for batch production of the ox‐carts that they were required to produce. Instead of providing facilities required to strengthen these workshops, the government was going ahead through SIDO (Small Industries Development Organization) to set up common facility workshops for industrial estates in various regions. The link (p.297) between R. & D. and manufacturing activities remained rather weak, making it difficult for programmes initiated by TAMTU to proceed beyond stage (3) or (4).

From the above observations, it appears that the link between the needs of the farmers (especially the small farmers) and the local manufacture of the kind of agricultural equipment they need has been inhibited by both weak R. & D. and weak base of local industry.

(d) Demand for agricultural equipment

The interaction between R. & D. activities and the local manufacturing capacity does not imply that there are no problems on the demand side. In fact there are considerable problems on the demand side, such as small size of national markets, an absence of subregional trading, general insolvency of the rural population (aggravated by inadequate agricultural credit, low producer prices, and restricted subsidies for equipment purchases), market instability resulting from policy and climatic changes, and competition from imports.

In most countries of Africa there is no clear agro‐mechanization policy (Commonwealth Secretariat 1977). This contributes to making it difficult to marshal such support measures within the overall development plans of these countries.

There is a need to stimulate demand for equipment in order to facilitate widespread adoption and expanded local manufacture of the equipment. However, credit and servicing facilities have been a constraint in some countries. For example, the Commonwealth Secretariat (1979 b) has reported that in Swaziland the Tinkabi (a small locally developed tractor) is selling in very small numbers (only 23 units were sold within the country in 1977 compared to the production capacity of 2,500 units a year and imports of about 200 tractors per year) because of lack of market promotion and credit facilities. It is reported that it is easier for a Swaziland farmer to get a bank loan of E7,000 for an imported Massey‐Ferguson tractor with no implements than a loan of E2,700 for a Tinkabi with a full set of implements (plough, planter, harrow, ridger, and trailer) built in. This is in spite of the fact that services for Tinkabi are well developed in the country and that farmers are keen to buy the machine, using it as a mobile maize mill and irrigation pump as well as for transportation, in addition to land preparation and planting. In his study of the development of ox‐cultivation in Tanzania, Kjaerby (1983) found that government actions tended to support tractorization via credit and allocation to villages and state farms. For instance, during 1981/2 TRDB (Tanzania Rural Development Bank),6 which is the main institution providing credit facilities for the rural areas, approved shs 7 million for tractors compared to shs 1 million for ploughs. Although oxen and ploughs are complementary there was no credit (p.298) facility for oxen. Kjaerby further observed that although donor reports by FAO (1975) and World Bank (1977) endorsed the superiority of the plough over the tractor, in practice they continued to channel their support facilities into tractorization.

On credit, the UNIDO (1983 a) study found that when Senegal halted distribution of credit to the co‐operatives there was a major collapse of demand for agricultural machinery. In contrast, Togo introduced credit for purchasing draught cattle resulting in increase in demand for animal‐drawn equipment.

As regards the competition from imports, the case‐studies (UNIDO 1983 a) have indicated that local production of agricultural tools in Kenya and Nigeria declined as a result of this competition. This suggests that there is a case for pursuing import policies which complement rather than inhibit local production.

These observations suggest that there is need to stimulate demand through

  1. 1. improving credit facilities especially to the small farmers;

  2. 2. improving service facilities for agricultural equipment, probably by establishing rural workshops in the areas where such equipment is used;

  3. 3. formulating mechanization policies which incorporate the interests and ability of the low‐income small farmers;

  4. 4. formulating and implementing import policies which enhance the complementarity between imports and local industry, especially in the field of production of agricultural equipment.

8.5. The role of industry in agricultural processing

Agricultural processing can play the role of stabilizing the incomes of peasants, preparing products for exports, facilitating inter‐regional and inter‐regional trade, stabilizing prices of food products between seasons, inducing an increase in marketable surplus and productivity by reducing wastage of seasonal surpluses and by extending the marketability of the farmers' crops nationally or internationally, and integrating agricultural and industrial operations. The share of Africa in world manufacturing value added (MVA) declined from 1.9 per cent in 1970 to 1.7 per cent in 1981 for food products, while for beverages and tobacco it increased respectively from 1.8 per cent to 3.1 per cent and from 2.9 per cent to 3.5 per cent (UNCTAD 1985: 4).

The state of the agro‐food sector in Africa reflects the relatively low levels of industrialization and the slow progress of agricultural production. In sub‐Saharan Africa the best results seem to occur in countries which have a diversified agricultural sector (e.g. Ivory Coast, Kenya, Zimbabwe, Cameroon, Nigeria). At the top of the scale are a few export industries operated mainly by foreign firms, usually TNCs. This extensive involvement of TNCs (p.299) in processing activities in Africa is influenced by factors such as the size of the domestic market and the development of export markets. Overall, however, the performance of the export agro‐industrial sector has not been impressive in spite of the good market networks that foreign firms are expected to have.

The developed market‐economy countries have tended to divert their imports of agro‐industrial products away from sub‐Saharan Africa, contributing to the decline of the share of imports from Africa from 8.7 per cent in 1974 to 4.3 per cent in 1984. The most sharply declining shares were those of food products (from 14.6 per cent to 8.2 per cent), processed vegetable and animal oils (from 21.7 per cent to 6.3 per cent), and wood products (from 13 per cent to 6.4 per cent) (UNCTAD 1985). Another factor which has contributed to this decline is the fact that developed countries that were previously major importers of agro‐industrial products have become self‐sufficient (or even surplus producers). Demand problems (e.g. low‐income elasticity of demand, utilization of synthetic substitutes) and protectionism (with tariffs escalating with the stage of processing) have contributed to the unfavourable trend. For instance, during 1970–80 industrialized countries almost doubled production of sugar even though beet sugar cannot compete with tropical sugar without protection. Yet in spite of the bleak prospects of sugar exports some African countries have invested heavily in sugar complexes (e.g. Ivory Coast, Cameroon, Sudan).

The dependence of agro‐industries on the agricultural sector for the supply of raw materials meant that they could not perform well when agricultural sector performance was low. For instance, in Somalià the low capacity utilization problem in the food‐processing sector has been attributed to the bottlenecks in agricultural production (UNIDO 1983 b). In contrast, Malawi has been pursuing a strategy of resource‐based industrial development where about 51 per cent of manufacturing value added (MVA) originates from agro‐processing. During 1973–9 about 70 per cent of the increase in MVA was in agro‐processing, reflecting the expansion of agriculture (UNIDO 1983 c).

From the technological point of view, the bulk of agricultural processing activity has taken place in large‐scale centralized production capacities. These technologies have been employed by foreign firms or state enterprises and have been associated with the modern farming sector rather than the small farmers who are in the majority. The logistics involved in having to supply raw materials to a few centralized large plants (e.g. in terms of infrastructure and maintenance requirements) have tended to erode the expected superiority of these enterprises. However, the local manufacture of equipment for the agro‐processing industry has remained virtually non‐existent. The bulk of agro‐processing activities have had a weak link with agricultural activities of small farmers on the one hand and with local industrial and technological activities on the other.

The positive contribution of local industry food processing has been inhibited by the choice of technologies which make use of imported capital intensive (p.300) plants, sometimes not based on the use of locally available resources. In addition, the organization of production is so centralized that the links between processing plants and the rural areas have tended to remain weak. For instance, Jones (1986) has argued that many of the decisions on food technology in Tanzania have not taken into account the balance of resources available locally. He has cited the case of National Milling Corporation's (NMC) large‐scale bakery which involved high capital cost in foreign exchange to process a product which is not adequately produced locally, hence not promoting food security based on local agriculture. On the latter point Andrae and Beckman (1985) have made at least two further observations in the case of Nigeria. First, they observe that the bread industry almost entirely depends on imports. Second, the presence of wheat flour mills obstructs any attempt to scale down wheat imports, and dependence on imported food is built into the industrial structure of the economy.

Jones (1986) has cited another NMC project, the Silo project, which is high capital cost, import intensive, and centralized. He argues that the same storage capacity could have been created at about one‐quarter of the cost (hardly involving any foreign exchange) by constructing 120 go‐downs spread throughout the country. In addition this would have improved food security in the regions, employment, the use of locally available construction materials, and would have economized on transport requirements. The decision to adopt this project was partly influenced by the already established large‐scale grain mills. Similar issues arise in the case of fruit canning, milk preservation, oilseed extraction, sugar production, and grain milling. In many cases, in practice, it is not a matter of choice between large‐ and small‐scale or centralized and decentralized plants, but of an appropriate combination of these reflecting different levels from village and district levels to regional, national, and subregional levels. What is rather conspicuous here is the very rare occurrence of decentralized organization of production at village and district levels.

It has also been observed that the existence of advanced technology for processing imported foodstuffs like wheat and barley has inhibited the development of technologies for processing locally grown food products. This has been demonstrated quite well by Andrae and Beckman (1985) in their study of wheat in Nigeria.

The observations made above suggest that in order to enhance the positive contribution of industry to the availability of and access to food, there is a need to pay greater attention to the decisions on food technology with a view to spreading the processing units to the countryside, to stimulating the local machine‐building industry, to economizing on infrastructural requirements (e.g. transport, complex management systems), and to stimulating employment and incomes in the economy particularly in the countryside.

(p.301) 8.6. Conclusion

This chapter has examined the positive contribution of industry in alleviating the food problem in sub‐Saharan Africa. It has been pointed out that the declining rate of growth of food production and falling food output per capita in Africa represent a threat of increasing hunger. The declining food self‐sufficiency ratios raise concern but it has been argued that food self‐sufficiency objectives must take into account country‐specific long‐term comparative advantages. The role of industry has been examined in terms of its contribution both to raising productivity in agriculture and to generating incomes in non‐agricultural activities.

It has been shown that only a small proportion of total investment is allocated to agriculture, the bulk of which goes to the large‐scale and usually capital intensive farms. The majority of the small‐scale farmers are bypassed by investments made in agriculture. It is suggested that in order to enhance productivity and entitlements of the majority of the low‐income small farmers, investments should be channelled to the smallholders' farm sector not only directly in agriculture but also in related activities like rural industry, rural workshops, and rural transport.

The chapter has indicated that non‐agricultural activities in the rural areas are important in making use of surplus labour (all the time or during the dry season when there are no farming activities), supplementing rural incomes, and reducing the risk of loss of entitlements on the part of the low‐income small farmers.

The chapter proceeded to observe that direct industrial employment has been limited largely to undertaking capital intensive investments. The relationship between large‐scale industry and rural small industry has tended to be competitive rather than complementary. This competitive relationship has resulted in loss of entitlements on the part of income earners in small‐scale industries. This suggests that there is need to make rural small‐scale industries and large‐scale industries complementary and create linkages between industry and food production activities in agriculture.

In respect of agricultural technology, an investigation of the case of agricultural equipment has indicated that the industrial sector in Africa produces a very small share of all equipment actually used in agriculture. Links between the existing basic industrial installations and agricultural machinery companies are very limited. The bulk of agricultural equipment is imported and there is a proliferation of the importing channels. In addition, there is little R. & D. capability at the producer level and this inhibits technological progress, and design and adaptation of equipment to local needs. The link between the needs of farmers (especially the small farmers) and the local manufacture of the needed agricultural equipment has been inhibited by weak R. & D. and a low base of local manufacturing industry. These findings suggest that there is a need to strengthen the link between R. & D. and small farmers on the one hand (p.302) and R. & D. and local manufacture on the other. Considering that the incomes of most small farmers are low, the role of credit is of considerable importance in generating an adequate demand for agricultural equipment. As regards imports, it will be appropriate to design import policies which are supportive of the local production of agricultural equipment.

The level of development of food‐processing industries in Africa is relatively low, reflecting low levels of industrialization and the slow progress of agricultural production. The bulk of food processing has taken place in large‐scale centralized production capacities mainly associated with the large‐scale farming sector. These activities have exhibited a weak link with agricultural activities of small farmers on the one hand and a weak link with the local industrial and technological activities on the other. This suggests that there is a need to pay greater attention to the development of local food‐processing technology, decentralize food‐processing units to the food‐producing countryside, stimulate the local machine‐building industry, and economize on infrastructural requirements.

(p.303) References

Bibliography references:

African Emergency Report No. 7, Apr.–May 1986, ‘A Periodic Report on the Crisis’.

Amani, H. K. R. (1986), ‘Food Problem: The Role of the National Food Strategy’, paper presented at the International Conference on the Arusha Declaration, Arusha, 16–19 Dec.

Andrae, G., and Beckman, B. (1985), The Wheat Trap: Bread and Underdevelopment in Nigeria (London: Zed).

Bates, R. (1981), Markets and States in Tropical Africa (Berkeley, Calif.: University of California Press).

Beeney, J. M. (1975), ‘Agricultural Mechanization Study’, UNDP Report to the United Republic of Tanzania (Rome: FAO).

Berry, S. S. (1984), ‘The Food Crisis and Agrarian Change in Africa: A Review Essay’, African Studies Review, 27.

Boesen, J., and Mohele, T. (1979), The ‘Success Story’ of Peasant Tobacco Production in Tanzania (Uppsala: SIAS).

Braun, J. von, and Kennedy, E. (1986), ‘Commercialization of Subsistance Agriculture: Income and Nutritional Effects of Developing Countries’, Working Paper (Washington, DC: IFPRI).

Burki, S. J. (1985), ‘The African Food Crisis: Looking Beyond the Emergency’, paper presented at a Conference on South–South Cooperation, 11–14 Nov., Harare.

Byerlee, D., et al. (1983), ‘Employment Output Conflicts, Factor‐Price Distortions and Choice of Technique: Empirical Results from Sierra Leone’, Economic Development and Cultural Change, 31.

Cliffe, L. (1978), ‘Labour Migration and Peasant Differentiation: The Zambia Case’, Journal of Peasant Studies, 5.

Collier, P., and Lal, D. (1980), ‘Poverty and Growth in Kenya’, World Bank Staff Paper No. 389 (Washington, DC: World Bank).

Commonwealth Secretariat (1977), Report of the Rural Technology Meet for East, Central and Southern Africa, Arusha, Tanzania, 29 Aug.–9 Sept.

———(1979a), Proceedings of the Appropriate Tillage Workshop, IAR, Zaria, Nigeria, 16–20 Jan.

———(1979b), Proceedings of the Appropriate Technology Coordinators Meeting, Lusaka, Zambia, 9–25 June.

Eicher, C. K. (1986), ‘Transforming African Agriculture’, Hunger Project Papers, 4 (Washington, DC).

Ellman, A., Mackay, B., and Moody, T. (1981), Guide to Technology Transfer in East, Central and Southern Africa (London: Commonwealth Secretariat).

FAO.(1975), ‘Assistance to Agricultural Mechanization: Tanzania, Mission Report’ (Rome: FAO).

———(1981), ‘Agriculture: Toward 2000’, Economic and Social Development Series No. 23 (Rome: FAO).

———(1983), Production Year Book 1981 (Rome: FAO).

Geofrey, M. (1985), ‘Trade and Exchange Rate Policy? A Further Contribution to the Debate’, in Rose (1985).

Goody, J. (1980), ‘Rice Burning and the Green Revolution in Northern Ghana’, Journal of Development Studies, 16.

(p.304) Heyer, J., et al. (1981), Rural Development in Tropical Africa (New York).

Hill, F. (1977), ‘Experiments with a Public Sector Peasantry’, African Studies Review, 20.

Hill, P. (1972), Rural Hausa (Cambridge: Cambridge University Press).

International Food Policy Research Institute (1986), International Food Policy Research Institute Report 8 (Washington, DC: IFPRI).

Ishikawa, S. (1964), Conditions for Agricultural Development in Developing Asian Countries (Tokyo: Committee for the Translation of Economic Studies).

Jaeger, W. K. (1985), ‘Agricultural Mechanization: The Economics of Animal Traction in Burkina Faso’, unpublished Ph.D. dissertation, Stanford University.

Johnston, B. (1951), ‘Agricultural Productivity and Economic Development in Japan’, Journal of Political Economy, 59.

———and Kilby, P. (1975), Agriculture and Transformation (Oxford).

———and Mellor, J. W. (eds.) (1966), The Economics of Agricultural Development (Ithaca, NY: Cornell University Press).

Jones, J. V. S. (1986), ‘Food Security and Economic Development in Tanzania? Past Problems and Proposals for a New Strategy’, paper presented to the International Conference on the Arusha Declaration held in Arusha, 16–19 Dec.

Kaul, R. N. (1979), ‘Some Considerations for Tillage Research’, in Commonwealth Secretariat (1979a).

Kigoda, M. A. (1984), ‘Incentives for Food Crop Production in Tanzania with Special Reference to the Mbeya Region’, unpublished Ph.D. dissertation, Columbia, University of Missouri.

Kjaerby, F. (1983), ‘Problems and Contradictions in the Development of Ox‐Cultivation in Tanzania’, Research Report No. 66 (Uppsala: Scandinavian Institute for African Studies).

Kongstad, P. (1980), ‘Kenya: Industrial Growth or Industrial Development’, in Rweyemamu (1980).

Lemma, H. (1985), ‘The Politics of Famine in Ethiopia’, Review of African Political Economy, Aug.

Lewis, W. A. (1954), ‘Economic Development with Unlimited Supplies of Labour’, Manchester School, 2, May.

Lipton, M. (1985), ‘Research and the Design of a Policy Frame for Agriculture’ in Rose (1985).

Mellor, J. W. (1986), ‘Agriculture on the Road to Industrialization’ in Lewis, J. P., and Kallab, V. (eds.), Development Strategies Reconsidered, US‐Third World Policy Perspectives No. 5 (Washington, DC).

———and Lele, U. (1973), ‘Growth Linkages of the New Food Grain Technologies’, Indian Journal of Agricultural Economics, 1.

Mettrick, H. (1978), ‘Oxenization in the Gambia: An Evaluation’ (London: Ministry of Overseas Development).

Mohammed, D. (1980), ‘Industrialization and Income Distribution in Ethiopia’ in Rweyemamu (1980).

Morrisson, C. (1985), ‘Agricultural Production and Government Policy in Burkina Faso and Mali’, in Rose (1985).

Muller, J. (1978), ‘Promotion of Manufactures of Rural Implements in the United Republic of Tanzania’, in UNIDO, Industrialization and Rural Development (New York: United Nations).

(p.305) Njonjo, A. V. (1981), ‘The Kenya Peasantry: A Reassessment’, Review of African Political Economy, 20.

Norman, D., et al. (1979), Technical Change and the Small Farmer in Hausaland, N. Nigeria (Department of Economics, Michigan State University).

O'Brien, J. (1985), ‘Sowing the Seeds of Famine: The Political Economy of Food Deficits in Sudan’, Review of African Political Economy, Aug.

Ohkawa, K. (1972), Differential Structure and Agriculture: Essays on Dualistic Growth (Tokyo: Hitotsubashi University).

Pearson, S., et al. (1981), Rice in West Africa: Policy and Economics (SUP).

Prattley, W. (1986), the special representative of the UN Secretary General in the Sudan quoted in Africa Emergency Report No. 7, Apr.–May.

Prothero, R. (1975), Migrant Labour from Sokoto Province, Northern Nigeria (Liverpool: Department of Geography).

Ranger, T. (1971), Agricultural History of Zambia (RLI).

Richards, P. (1983), ‘Farming System and Agrarian Change in West Africa’, Progress in Human Geography, 7.

Rose, Tore (ed.) (1985), Crisis and Recovery in Sub‐Saharan Africa (Paris: OECD).

Rweyemamu, J. F. (1980) (ed.), Industrialization and Income Distribution in Africa (Dakar: CODESRIA Book Series).

Samoff, J. (1980), ‘Underdevelopment and its Grass Roots in Africa’, Canadian Journal of African Studies, 14.

Schoepf, B. G. (1985), ‘Food Crisis and Class Formation in Shaba’, Review of African Political Economy, Aug.

Sen, A. K. (1981), Poverty and Famines (Oxford: Oxford University Press).

———(1986), ‘Food, Economics and Entitlements’, Working Paper 1 (Helsinki: WIDER).

Shepherd, A. (1981), ‘Agrarian Change in Northern Ghana’, in Heyer et al. (1981).

Singh, I., Squire, L., and Strauss, J. (1986), ‘A Survey of Agricultural Household Models: Recent Findings and Policy Implications’, World Bank Economic Review, 1.

Spencer, D. S. R. (1986), ‘Agricultural Research in Sub‐Saharan Africa: Using the Lessons of the Past to Develop a Strategy for the Future’ in Berg, R. J., and Whitaker, J. S. (eds.), Strategies for African Development (Berkeley, Calif.: University of California Press).

United Nations (1983), Year Book on National Accounts Statistics 1981 (New York: UN).

UNCTAD.(1985), ‘Protectionism and Structural Adjustment: Problems of Agro‐industrial Production and Trade’, TD/B/1086, 23 Dec., Geneva.

UNIDO.(1983a), ‘Agricultural Machinery and Rural Equipment in Africa: A New Approach to the Growing Crisis’, Sectoral Studies No. 1 UNIDO/IS.377, 25 Mar., Vienna.

———(1983b), ‘The Potential for Resource Based Industrial Development in Least Development Countries’, No. 6 Somalia UNIDO/IS426, 16 Dec., Vienna.

———(1983c), ‘The Potential for RBID in LDCs’, No. 5 Malawi UNIDO/IS 389, 15 June, Vienna.

———(1985), Industrial Development Review Series, Nigeria, UNIDO/IS 557, 9 Sept., Vienna.

Venema, L. B. (1980), ‘Male and Female Farming Systems and Agricultural Intensification in West Africa: The Case of the Wolof, Senegal’, in Presvelon, U., and (p.306) Spijkers‐Zwart, S. (eds.), The Household, Women and Agricultural Development, Miscellaneous Papers 17 (Wagemingen: Land‐Labour Hogeschool).

World Bank (1977), ‘Rural Development Sector Policy Paper (Tanzania)’ (Washington, DC: World Bank).

———(1981), ‘Accelerated Development in Sub‐Saharan Africa’ (Washington, DC: World Bank).

———(1984), ‘Toward Sustained Development in Sub‐Saharan Africa: A Joint Programme of Action’ (Washington, DC: World Bank).

World Food Council (1984), ‘Food Strategies in Africa’, Progress and Critical Issues, June.

Wright, D. (1977), ‘Integrated Programme for Technical Change’, in Commonwealth Secretariat (1977).


I am grateful for useful comments made on an earlier draft by a number of colleagues. Specifically I would like personally to acknowledge the insightful comments of Paul Streeten who kindly accepted to be a discussant of the first draft of this paper at the WIDER Conference. Comments made by Amartya Sen, Carl Eicher, Barbara Harriss, Judith Heyer, and Nanak Kakwani have contributed considerably to the improvements made in this chapter. However, as usual I take responsibility for the views expressed in it.

(1) The evidence is discussed in greater detail in Jean‐Philippe Platteau's contribution to the second volume of this book.

(2) The concept of self‐sufficiency as used by these sources (e.g. FAO, World Food Council) only captures the relationship between domestic food supply and food imports (including food aid) and does not reflect the adequacy of food supply (e.g. in terms of nutrition).

(3) The study by UNIDO (1983 a) is based on a major data‐gathering exercise which was undertaken in 16 countries of Africa in 1981. Each case‐study focused on the ability of the country to produce agricultural hand‐tools, machinery, and equipment locally and on the demand for such goods at the beginning of the 1980s. The 16 countries represent different development levels and agroecological conditions in the four geographical subregions of Africa. In this sample agriculture contributes 35% of the GDP and employs 74% of the active population.

(4) The world average is 55% compared to 60–70% for Africa.

(5) The workshop was organized by the Commonwealth Secretariat in Zaria (Nigeria) in 1979.

(6) It has changed its name to CRDB (Cooperative and Rural Development Bank).