Smallholder responses to
globalization: African field
experiences

John Dixon, Aysen Tanyeri-Abur and Horst Wattenbach

Introduction^[3]

This chapter is based on field experience of the FAO-World Bank Study of Globalization, Structural Adjustment and Smallholders, for which field work has been conducted in Ghana and Mali in West Africa and Uganda and the United Republic of Tanzania in East Africa. These countries have experienced moderate to rapid economic reform^[4] and thus significant effects of globalization on smallholders might be expected. For this reason, the responses to globalization identified in these countries illustrate the types of smallholder adjustments which could be expected in similar conditions elsewhere in Africa and other regions.

For the field research, three populous and economically important farming system zones with different levels of market access (from moderate to well integrated) were selected within each country (see Table 1). In each zone typical sites were identified for the field research. This design allows the comparison of impacts across different levels of market access within one country, and also the comparison of impacts on similar commodities/systems across different national policy environments. Whilst all sites produce food crops, maize is the staple in seven sites. Six sites produce traditional export crops: four sites are located in tree crop based farming systems (cocoa or coffee), and two sites produce cotton. Three sites are agro-pastoral systems, based on sorghum, millet and livestock.

Each research site comprised a cluster of a principal research village (or community) and two nearby subsidiary villages. Within each village or community, at least three different types of farm households were identified according to household food security status or wealth. Thus, a spectrum of household types was differentiated, including medium and large farms and non-agricultural households where they existed. In each village a participatory rural appraisal was conducted by a multi-disciplinary team. In the principal village in each farming system, a household interview survey (of 40 households) was conducted.

Table 1: Study sites

Note: FS farming system (see Annex for equivalences to FAO-World Bank African regional farming systems, Dixon et al 2001)

Country context

In order to facilitate the interpretation of the field survey findings, the key aspects of the macro-economic and sectoral context are summarized in Table 2. Gross national income per capita lies in the range of US$210-290, poverty is extensive (20-73%), agriculture accounts for 35-45% of GDP and trade is substantial (exports 10-49% of GDP; imports 23-70% of GDP).

At the same time, they show marked differences in a number of other economic indicators. The GNI per capita measured in PPP terms ranges from $540 in the United Republic of Tanzania to $1980 in Ghana. Although around two-thirds or more of the population of all four countries live in rural areas, Uganda shows the lowest degree of urbanization. Overall, foreign trade of the two West African countries is considerably higher than for the two East African countries, with Uganda having the lowest ratio of exports to GDP. The United Republic of Tanzania has the lowest prevalence of poverty; Mali the highest (73% poor with less than US$ 1/day). Within countries, neither poverty nor smallholders are uniformly distributed; for example, in Uganda, the south-west and south-east are dominated by smallholders.

Table 2: Selected national economic indicators

Source: *=World Development Report 2003; **=World Development Indicators 2002

All four countries are moderately to well advanced on economic reforms. Table 3 presents some key milestones of economic reforms in the four selected countries. Economic reforms commenced in 1983 in Ghana and in the 1985-87 period in the other three countries. The sequence and pace of implementation of reforms varies considerably across the four countries.

Uganda is sometimes referred to as a particularly striking example of the speed of recovery in the wake of substantial economic reforms (see also Brett 1996) and thus it is chosen to illustrate certain aspects related to the impact of globalization on smallholders. The main export crop coffee was under strong state control and farm gate prices fell rapidly during the 1970s (by 1981 prices had fallen to one quarter of the 1972 level). The first economic reform program was launched in 1987, foreign exchange controls were lifted in 1989, and in 1990 the Uganda Coffee Marketing Board (CMB) monopoly was weakened by allowing export by Union Export Services (UNEX). Two years later, the further liberalization of coffee marketing stimulated a strong supply response from producers and vigorous entry of private exporters who rapidly seized market share from the state sector (Dijkstra and van Donge 2001). Among other advantages of the private sector marketing, farmers received cash payments on delivery instead of the promissory notes issued by the state bodies. The coffee marketing system survived the end of the coffee boom in 1996 largely intact, although there was some concentration of exporters. The rising share of the farm gate price in export prices (from 20% in 1990/91 through 60% in 1995/96 to 82% in 1997) buffered farmers from the slump in international prices after 1996. They also benefited from the premium on Robusta coffee because of its suitability for instant coffee. However, the premium may be threatened by declining quality of exports as cash-strapped farmers deliver unripe and poorly dried coffee.

Table 3: Key economic and trade reform milestones

Note: the implementation of some reform measures occurs over several years from the start year shown in this table

The Ugandan cotton industry contracted after 1973/74, reaching a low in 1987/88 (prices had fallen by 1981 to one sixth of the 1972 level, and cross-border sales had increased). Initial liberalization attempts failed to stimulate production (but drought and conflict may have impeded recovery). These early attempts were based on re-capitalizing the co-operative unions, but the capital injections were lost through mismanagement. The Lint Marketing Board retained a monopoly until 1993/94. Complete liberalization of the sub-sector generated a strong supply response in 1996/97 (Dijkstra and van Donge 2001). Non-market factors have been of particular importance in the case of Ugandan cotton; seed quality, fertilizer and insecticide availability and crop finance have dampened supply response; and some exporters who financed production inputs failed to recover their loans when farmers sold to competitors. Given the cash requirements of cotton and the poor availability of crop finance, cotton has not become a small-holder crop in Uganda.

Whilst coffee exports declined in importance non-traditional exports grew dramatically, including fish exports from Lake Victoria, as well as the expanding cut flower industry, fresh fruit and vegetables and essential oils and spices (mostly vanilla). The driving forces differed. The expansion of fishing has been driven mostly by strong international demand combined with the liberalization of the sub-sector; and substantial employment is being generated on the shores of Lake Victoria and for middlemen, even though the export trade as such is mostly in foreign hands. The flower industry developed partly as a pre-emptive move of importers to secure supply in the wake of expected disruptions of existing sources of supply in other countries (Dijkstra and van Donge 2001); and its sustainability may be threatened by the recovery of competitiveness in other countries including Kenya.

In contrast to many other African countries, Ugandan state intervention in the food crop sector had been minimal. Food crop producers, however, experienced modest income improvements during the reform period due to the dramatic falls in cash crop prices which led to a shift from cash to food crops. The loss of draught animals and the deterioration of rural infrastructure during the civil strife in the pre-reform period were overall factors that hindered the ability of farmers to respond to reforms.

The broad outcome of the reforms is reflected in the distribution of rural poverty. The prevalence of poverty is highest in the dryer and remote northern parts of the country (almost 68% of the rural population), compared to only 30-35 percent among the central and western rural population. Appleton attributes about half of the poverty reduction up to 1996/97 to coffee-growing households (Appleton 2001; Appleton et al 1999). The availability of consumption items might have improved as a consequence of trade liberalization, but at the expense of price effects for consumer goods (Gibbon et al 1993).

Price transmission to farmers

If agricultural price signals are not transmitted to farmers then, ipso facto, trade liberalization will have no direct effect on smallholder production and consumption behaviour (although indirect responses could still arise through factor market effects or improved institutional arrangements). Among the four case study countries, two were deliberately selected to represent landlocked countries. Consequently, the transport cost ‘wedge’ depresses the export potential of locally produced tradable goods while at the same time providing natural protection for locally produced tradables. The cost of fertilizer and other agricultural imports increase significantly. Where the existence of time series data on local commodity prices in relevant markets permits, price transmission effects will be quantified; and during the second phase in depth analysis is planned of selected commodity chains. Meanwhile, pending quantitative analysis based on time series market data, the perceptions of farmers and local key informants concerning the extent of price transmission is presented in this section.

As indicated in Table 4, price transmission to farmers has been higher in areas of good market access, as expected. Conversely, in the sites with relatively less market access farmers perceive that the degree of price transmission is less and takes longer. With regard to traditional export commodities, in Ghana the changes in world price of cocoa appear to be transmitted to farmers. A high transmission of international price signals was also noted for coffee in the United Republic of Tanzania (Arusha) and Uganda (Mukono). In both countries the removal of the guaranteed price for coffee exposed producers fully to the international markets in the classic sense. As proxy-indicators of the price trends for different commodity groups since liberalization, farmers were asked to compare the terms of trade of their basic purchased items, such as paraffin, soap and salt to their main cash crop.

Table 4: Extent of price transmission

According to their interpretation, coffee prices in Uganda (Luwero) failed to keep up with the increases in the prices of their main consumer goods. No price series of local prices for the other commodities in questions are as yet available to abstract price developments from the decline in coffee prices. In the case of cotton in Mali there is no direct market-articulated transmission of international prices to producers: 95 percent of the production of cotton is exported (controlled by the CMDT) under administered price arrangements.

The situation for food crops is rather more complex. Generally price transmission is lower than for export crops, but domestic production faces imports especially in Ghana and the United Republic of Tanzania. Location and season influence the extent of price transmission for maize in the United Republic of Tanzania, where harvest prices especially may be influenced by the costs of imports to urban areas; and also the prices in neighbouring countries. However, prices soar later in the season suggesting a failure of storage.

In livestock, cheap imports have adversely affected local markets. The agro-pastoral areas have been hit repeatedly during the adjustment period by falling livestock prices caused by cheap imports of non-African meat, indicating that price transmission within Mali has worked, even though to the detriment of the population in the north of the country. In Ghana chicken, pork and turkey parts are imported and the poor prefer these items due to lower price.

Responses of smallholders

The adjustment processes within the farming systems in the four countries discussed above is illustrated in Table 5. The response of farmers to the price transmission is not solely based on price effects, as their adjustment possibilities are determined by a large number of complementary driving forces. Among those is the level of fixed assets of the farming system, most strongly on the coffee systems of the study areas. Technical knowledge to adapt to emerging opportunities, demographic factors as well as possibilities to obtain crop finance through formal or informal sources are other key factors.

In the cotton farming systems zone of Mali, field teams observed contrasting tendencies in the cropping patterns, indicating an expansion of cotton and maize areas for larger farms at the expense of sorghum and millet, while small farms show a contrary reaction. This heavy expansion of areas and production of cotton and of dry cereals coincided, however, with no increase in production intensity. Imported fertilizer was partly replaced by organic manure due to sharply increased fertilizer prices after devaluation. In the cotton zone, large farms generated sufficient income to increase herd sizes and intensify breeding.

Table 5: Direct adjustments on cropping patterns and input use

Within the cash crop systems, the case studies show interesting differences in response to increasing fertilizer prices: whereas the higher producer prices for cotton caused a rapid expansion of cotton areas since 1994/95, yields have remained stable. According to farmers, cotton yields are threatened with further decline as input levels are being reduced to cut production costs: average fertilizer use on cotton has been reduced since 1992 from approximately 150 kg/ha to 100 kg/ha. Given the disengagement of the CMDT, combined with a lack of commercial crop financing possibilities, this could lead to a deterioration of the cotton sub-sector. Especially poorer farmers are being affected by this situation.

In the rice based farming systems zone (Office du Niger) of Mali, the reduced control by the state in the decisions on the development of the areas surrounding the irrigation schemes has led to the expansion of diversified cropping systems (e.g., shallot cultivation), as well as forward linkages to the rice processing industry (e.g., threshing). At the margins of the irrigated land, especially women and young people are reported to benefit from the improved profitability of rice farming, following the renewal of competitiveness after devaluation in 1994. Among the adaptation strategies are the double cropping of rice, increase in the amounts of manure applications, as well as the adoption of improved varieties and based on marketing considerations (quality), hire of equipment and informal leasing of land.

Livestock as the savings account of choice was noted across the farming systems zones of Mali, with stronger tendencies in the drier systems with sufficient grazing. Diversification of agricultural production was not a feasible option for most farmers residing in the agro-pastoral system of Mali (e.g., north of Mopti), representing the low market access site. The regained relative competitiveness due to devaluation of FCFA was used by herders to expand herd sizes, combined with growing exports towards Côte d’Ivoire and increasing importance of beef and sheep fattening for the market.

In the recent decade in Uganda, export opportunities have been increasingly being taken up by commercial farms in the most advantageous locations close to Kampala by specializing in coffee, tea, cocoa, flowers and sugarcane production. The latter are partly stimulated by the urban demand and destined for processing factories, such as a sugar plants and a soft-drink factory. The shift towards the expansion into new crops was also partly based on a push-factor, as traditional cash crops (coffee and banana) were affected by disease (wilt) and low prices (especially coffee). This resulted in a shift to modern cash crops, such as vanilla, pepper and tomatoes due to better prices and the readily available markets.

Some respondents were of the view that those who succeeded to diversify into vanilla production have benefited most from globalization and structural adjustment, although commodity price declines would inevitably threaten such good results. Women might have been particularly successful in doing so, but there was at the same time the risk that traditional and modern elites would gain a disproportionate share in the future.

The response of farming households to shocks like the slump in international coffee prices has been to reduce investment. This has led to reduced output per acre and reduced incomes. Other shocks like trade and price liberalization, institutional reforms leading to private sector provision of certain services, have not been adequately absorbed due to a limited resource base and farmers lacking credit to access privatized services. On top of these economic shocks, natural shocks such as the banana and coffee wilts have destroyed important parts coffee and banana plantations leading to reduced yields and eventual revenue.

The food crop production basket remained relatively stable, with the exception of banana due to pest and disease incidents. Adverse climatic events led to a decline in marketable surplus, and inputs are considered too expensive and difficult to access especially in remote locations. Yields are therefore generally considered as being on the decline. Hired labour increased (due to school enrolment). Households that practice mixed farming have increased since the 1980s, to nearly half of all households, and all farmers practice intercropping.

Fallow periods have been reduced from 3 years in 1980s to 6 months or none at all. Hired labour use has increased since the 1980s and piece work payments in cash rather than in kind are now common.

Farmers in the coffee-banana system of Uganda summarized their experiences with their farming strategies as follows: yields of major food crop are said to have halved in the last 20 years. The reasons for the declining trend change over the period under consideration, though: during the 80s, civil strive reduced attention to cultivation practices and scarcity of inputs, the 90s low produce prices, declining soil fertility due to the untimely supply of fertilizers, partly expired chemicals to treat against pests and diversion of inputs at the extension service level; recent shortcomings in the 2000s are quoted as being: inadequate access to farm machinery, information deficits in the extension service to deal with new crops and diseases as well as high input prices.

The mixed agro-pastoral systems zone of Uganda (e.g., Masindi), had received support by the state run tractor services. Due to their discontinuation, mechanization levels have declined and, conversely, ox-cultivation expanded in the 1990s. Hired labour is becoming increasingly important as children are less involved in farming due to increased school enrolment, and a general monetization of social relations is reported. At the cropping systems level, diversification has increased through the addition of cassava, potatoes and maize to the previously dominating millet, groundnuts and beans. Parallel to this, traditional foods crops are gaining the additional role of becoming cash crops. The role of cotton as a cash crop has declined, with the argument of lacking market outlets. However, farmers are contemplating of returning to cotton cultivation, which might have been stimulated by price information on newly established local radio stations.

The overall situation of structural adjustment response in the farming systems of Uganda is as follows: agriculture-oriented villages have not fully benefited from globalization and structural adjustment. In terms of production, there seems to be a shift of emphasis towards commercialized food crops like maize. Poverty continues to be widespread and liquidity for crop finance is a major concern. Another factor could be the limited capacity, particularly of small holders to recognize, explore and compete favourably under the new set up and opportunities, such as improved markets. Other reasons include inadequate infrastructure, lack still deficient market information, poor resource endowments (low access to more land), endemic pests and diseases as well as human diseases. Farmers failed to recover from the civil strife of 1980 - 1986 from loss of property.

Farmers have generally responded strongly to the price increases after disengagement of the public sector in supplying fertilizer to the price increases. Input use has, however, dropped less markedly in the low potential areas of the case study countries, but this is explained by the low initial levels of fertilizer use. The crop production systems, particularly in the semi-arid areas of Mali and Uganda, but to a similar degree in the case of Ghana and the United Republic of Tanzania, are dominated by sorghum or millet based systems under highly variable rainfall conditions, where the economics of fertilizer application is a sensitive one.

Village level effects

It was expected that different indirect or second round effects could be observed at the village level in the research sites. The field teams observed major changes with respect to financial services, monetization of the village economy, trade patterns as well as institutional change of public services. While the quantitative results are still being analysed, some indications on these second round effects from the PRA data are available.

Several parallel developments have led to an increased monetization of the village economies. Some of them are directly linked to trade liberalization, while others are caused by the privatization of public services. Among the latter, the field teams report increased cash needs at the household level to pay for school fees (Uganda), and the labour force lost by children attending school is being replaced by hired labour through the rural labour market. Labour relations have simultaneously shifted from payment in kind to almost exclusively cash-based arrangements.

The higher profitability of rice cultivation in the irrigated farming system in Mali has led to intensification of the production technology, and farmers rapidly accepted transplanting they had previously adapted only to a very low extent. This technological change in turn considerably improved the absorption of labour force which had fled from the north of the country in past droughts.

The privatization of the marketing system for cash crops increased accessibility of cash in the rural economy, as traders are paying in cash and at the farm gate for the produce. Trading intensified, partly caused by increasing availability (and affordability) of transport facilities for traders (motorbikes in the remoter parts of Uganda, and to some extent cars in the vicinity of Kampala). Women have particularly expanded their trade or entered with new businesses in Ghana, where women had a traditional position in trading. Farmers themselves increasingly use bicycles for transport, and their prevalence in the villages has sharply increased in Uganda in all study locations, where in some villages 70 percent of households now own a bicycle up from 20 percent in the 1980s. Deficient infrastructure remains a problem in order to reduce transport costs and to realize economies of scale in marketing, and the rural feeder road system has mostly been expanded in the districts neighbouring Kampala. Increased prevalence of bicycles and motorcycles - as Masindi - has made marketing easier, and due to the increasing prevalence of rural radio, price information has also become more widespread. In Uganda, even in the less accessible areas, mobile phone coverage had reached the medium integrated case study villages.

Processing of local agricultural and other products has increased across all systems and market access conditions in Ghana, even though due to higher population density and income levels the intensity of transactions is highest in the forest zone. In the medium and low access zone, cassava processing into gari increased, while in the low access area, shea butter demand and processing expanded. The equivalent processing activities in the medium access zone of Mali is based on milk although monopsonistic market structures and very low local demand for milk is an inhibiting factor for further development and income generation.

Table 6: Village level effects on smallholders

The lack of financial services is one of the major constraints in the farming community to benefit from the improved profitability of the cash crop sector, as input supplies are mostly provided by private traders, and liquidity constraints have prevented the intensification and yield increases in the cotton system of Mali. The rice sector, however, has seen remarkable intensification; fertilizer use has increased and yields have more than doubled in the last decade. Access to financial institutions has not improved in the last decade, and farmers continue not to rely on the only available bank in the district for their needs. However, the accessibility of micro-finance institutions, particularly for women, has improved. Farmers in Mali in both the cotton and the rice farming systems have invested in expanding their livestock herds as a savings deposit, or increasingly engaged in trade. Livestock ownership, particularly in the case of cattle, however, is reported to show an even higher bias towards the largest of the four farm size groups in Mali than cash income in general.

Increased commercialization and monetization of the rice industry has led to social change in the Office de Niger area. Traditionally, the head of extended households represented the interests of a number of nucleus households, including important allocation and production decisions. This structure has come under increasing pressure, as the interest of nucleus families to gain independence has markedly increased. Hence, the share of small farms in the farming community has considerably increased. Positive social change was reported for the low access zone of Mali, where the increased profitability of the livestock sector helped strengthening the breeder associations.

Linkages to input supply and produce traders are reported in all systems of Mali, and field studies confirm strong diversification of the trade relations through the expansion of small commerce, processing industry and small-scale milk processing. The agro-pastoral system in the north documents has seen increased cattle trade, which was mentioned as an employment possibility for young people in the north of the country. The increased profitability of cattle breeding has in turn led to a strengthening of co-operative associations and breeder co-operatives.

The withdrawal of the state in the provision of public services, such as health and education, is mostly seen as negative by the farming communities in the case study areas. In Uganda, and strongly expressed by the population in Luwero, is overwhelmingly of the opinion that health services have declined in quality, judged by the fact that drugs were more readily available and affordable than at present, while now the health service has been reduced to the prescription of drugs which farmers often fail to afford from the free market.

However, there are notable exceptions on the effects of the withdrawal of the state reported from the rice production system of Mali, where the increased rice yields have improved the conditions for village infrastructure development. Village associations maintain a virtual monopoly of rice threshing and their business and income has rapidly increased parallel to production. The proceeds from these activities are invested in village infrastructure, such as schools, health posts and rural electrification. However, not all of these associations achieved sufficient control over these funds to invest them wisely, and large farmers succeeded in some villages to get hold of resources to fund their farm inputs. In Mali, particularly in its rice and cotton based systems, the state has succeeded in providing funding for the expansion of health posts and add primary schools despite the general trend for devolution. It was also observed that as disposable income falls due to the stagnation and fall in the cotton sector, other sectors, including services in the village and surrounding areas are declining and resulting in poverty and destitution which results in migration to other cities. For example, Koutiala, the city in the cotton area was certainly on the decline with most of the hotels and restaurants and other business closed, and overall very little economic activity of any type present)

Among the most important institutional changes in Uganda over the past two decades respondents cited decentralization of the local administration, which allows the rural population closer interaction with and control over their Local Council (created in 1993), which receives 25 percent of the taxes paid by residents. Extension services continue to exist at the district level, but their mobility has decreased and farmers need to visit them in their offices for advice. Farmers’ interest in the service appears to have declined in general, as there is no associated input provision complementing recommendations. Input supply and the number of stockers at the district level are stated to have improved; however stockers are mostly found in the district town 30 km from the study villages. Fertilizer applications have been reduced, and in combination with pest pressure, farmers in Luwero - Uganda state that yield levels have declined. Farmers give the decline in the yields as the main reason why off-farm work is seen as more lucrative than farming, while it is then as the opposite for the past. Official extension services have declined in the past decades, but fortunately were partly replaced by alternate services by NGOs.

Essential consumption items in the rural areas are considered paraffin, salt and soap, which had affordable prices in the past, but were frequently unavailable. Even though prices have considerably increased since the 1980s, their availability even in this remote location is not a problem, even though farmers complain about the price levels. In terms of welfare indicators, the subjective assessment of the rural population is that health and school fees have increased more than income levels.

Although it is difficult at this stage to quantify the prevalence of households which have gained or lost from globalization in the study villages (or the level of gains and losses) it can be said that larger farmers benefited across the board to a higher degree from the improved potential for market integration. In Mali, the importance of crop finance to reap the benefit cotton market liberalization caused larger farms to benefit disproportionably from the higher share of world market prices of cotton paid to farmers. At the same time, a decline in the share of the largest farm group in the number of exploitations is reported, partly explained by the expansion of cotton cultivation into new areas and the general increase in the agricultural population.

Smallholder characteristics determining impacts

Specific characteristics of benefiting households in the cotton sector of Mali are the following. Well equipped large producers (15 - 20 persons and 15 - 20 ha average holding size) who own large cattle herd and who cultivate between 1/3 and over 50 percent of cultivated land with cotton. The equivalent characteristics in the rice production system are large and medium producers with more than 4 ha of irrigated land and 15 family members, which are well equipped and frequently own a threshing and/or shelling machine and/or cattle herd.

Table 7: Characteristics of ‘winners’

Within the agro-pastoral system, the capacity to invest in increased market opportunities is most pronounced in the group of large and medium agro-pastoral and pastoral producers with herd sizes above 30 heads for agro-pastorals and 50 to 100 heads for pastoralists.

Preliminary results from the Uganda field study suggest that training in business development provided a useful tool for households to benefit from the market opportunities. The most successful cases are however reported again for households with business experiences, frequently with the involvement of women. It is said that in the coffee-banana system of Mukono and Luwero District, the proportion of women with off-farm income increased from 10 percent in the 1980s to 80 percent today. Education is seen as playing a pivotal role for benefiting from commercial opportunities, which explains the readiness of farmers to invest in it, despite their complaints about the associated costs. Institutional relationships continue to play their role as well, though, as connections to the political or social elite is stated as beneficial to the development of new cash crop or private enterprises, including vanilla and flower cultivation.

Conclusions

Economic reforms since the mid-1980s have created an institutional environment which has led to profound, albeit uneven, impacts of trade liberalization at the household and village levels. Nevertheless, some common patterns can be tentatively identified from the preliminary results. The variation in the sequence and pace of reforms together with the particular characteristics of the countries and the farming systems have created a mosaic of different impacts across sub-Saharan Africa. The basic process underlying the local impacts of globalization is the transmission of prices from international markets to producers, which varied across commodities, research sites and countries, depending on the structure and performance of the market institutions for the different commodities and locations. As expected, communities perceived a higher degree of price transmission for a range of commodities in the sites with good market access. Transmission has been higher in the case of liberalized export crops such as coffee in Uganda and the United Republic of Tanzania but lower in the cases of regulated commodities. Price transmission has been significant for horticulture and in a number of cases for livestock and food crops, including maize, cassava and rice (including market effects of the imports of cheap food grains and EU meat and poultry parts). Transmission is affected by transport costs (a function of distance to markets and transport infrastructure) and the performance of market institutions.

Farmers adjust to such agricultural input and output price signals in a number of ways. Cash crop farmers, particularly in high market access areas have generally responded strongly to the price changes, often diversifying towards horticultural production. With the decline of traditional export commodity prices, smallholders have reduced coffee areas and shifted into new cash enterprises such as vanilla in Uganda and dairy in the United Republic of Tanzania - and there has been some shift from cash crops back to food crops. After disengagement of the public sector in supplying subsidized inputs, fertilizer use has dropped markedly in the low potential areas. The response of farmers to the price transmission is not solely based on price effects, as their adjustment possibilities are determined by a large number of complementary factors including infrastructure and transport costs (especially Mali and Uganda), lack of rural finance (especially low market access sites), lack of complementary inputs (such as animal traction in Uganda), information on diversification opportunities (all countries) and asset fixity.

Where inter-sectoral linkages are strong a variety of indirect or second round effects have also been observed. Where additional incomes are spent in local areas, the overall impact on development may be higher. Conversely, lower incomes from falling prices may have negative effects on other sectors that depend on local expenditures generated through incomes from agricultural production. An increased degree of monetization of the village economy has been observed (for example, the substitution of hired labour for family labour in Uganda), and generally increased off-farm income from petty trade and local businesses. Again for villages with higher market access and including farmers that have diversified into new export crops, the local economy in general and inter-sectoral linkages remain strong. In poorer regions, particularly affected by state withdrawal and stagnating agricultural activity, inter-sectoral linkages are weaker, resulting in increased poverty in the immediate and surrounding local areas and increased out migration. There are indications of increments to social capital in certain situations, for example stronger livestock breeders associations in Mopti, Mali. Conversely, elsewhere in Mali traditional farmer-pastoralist linkages have weakened. However, local institutional weaknesses may have constrained the growth of inter-sectoral linkages.

Among the communities in the research sites, the winners from globalization tend to be the rich and large farmers, where they have capital to invest and good access to information on new opportunities. The knowledge of new opportunities for diversification is derived through public media, family networks and market engagement. Given the weak rural finance systems, profitable cash enterprises and off-farm income are two major source of farm investment capital. In this respect, smallholders in some settings are more flexible because of less asset fixity. These findings point to the importance of information flows and the availability of capital to smallholders for them to take full advantage of the opportunities of globalization.

References

Appleton, S., Emwanu, T., Kagugube, J., & Muwonge, J. 1999. Changes in poverty in Uganda, 1992-1997. WPS/99.22 pp. 43. Oxford, UK: Centre for the Study of African Economies, University of Oxford.

Appleton, S. 2001. Changes in poverty and inequality. In Reinikka. R., and Collier. P., (Ed.), Uganda's recovery: The role of farms, firms, and government. Vol. 2 1994 pp. 83-122. The World Bank Washington D.C.

Brett, E. A. 1996. Uganda. In Engberg-Pedersen, P., Gibbon, P., Raikes, P., and Udsholt, L., (Ed.), Limits of adjustment in Africa: The effects of economic liberalization, 1986-94. pp. 309-46. Copenhagen, Denmark: Heinemann, USA.

Dijkstra, A. G., & van Donge, J. K. 2001. What does the 'show case' show? Evidence of and lessons from adjustment in Uganda. World Development 29(5), 841-63.

Dixon, J., Gulliver A., and Gibbon, D. 2001. Farming Systems and Poverty:: Improving Farmers’ Livelihoods in a Changing World. FAO and World Bank. Rome and Washington, DC.

Drafor, I., Kwadzo, G. & Fialor, S. 2003. Globalisation, structural adjustment and smallholders in Sub-Saharan Africa. Ghana report for FAO/World Bank.

Gibbon, P, Havnevik, K.J. & Hermele, K. 1993. A blighted harvest: The World Bank and African agriculture in the 1980s. London, UK: James Currey.

Hella, J., Mbiha, D., Nyange, D., & Makindara, J. Forthcoming. Globalisation and small holders in Tanzania. Tanzania report for FAO/World Bank.

Kebe, D., Bellieres, J.F., Sanogo, O., with Teme, B., Coulibaly, B., Djouara, H., Kone, Y., & Toure, S. M. 2003. Impact de la globalisation et de l’ajustement structurel sur les petits producteurs. Mali report for FAO/World Bank.

University of Makerere. Forthcoming. Uganda report for FAO/World Bank.