2.3.2 Institutions

2.3.3 Organisations

2.3.4 Politics

2.3.5 Culture

2.3.6 Society

2.3.7 Policy

4.2.1 Questions of immediate relevance to economists' projects

4.2.2 Prior sociological questions

4.3.2 Indigenous definitions of poverty:

4.4.1 Units of measurement: what are we measuring and why?

4.4.2 Defining levels of economic poverty:

The snapshot household survey identifies certain people as poor at a particular point in time. However, households and people move in and out of economic poverty throughout their lives. This raises issues of vulnerability and variability, rural differentiation, the inter-generational transmission of deprivations/advantages, and the correlation and convertibility of poverty dimensions (for more examples from Uganda see Appendix 2).

Vulnerability may result from living in a risky environment, or from living on the edge of destitution, where one piece of bad luck can start a downward spiral. Disaster may strike individuals (and their households) or whole communities or regions. Recently parts of Uganda have suffered from famine, earthquakes, evictions, physical insecurity, cattle-raiding and other forms of banditry, water hyacinth etc.

Variability: People tend to be poorer at certain times in the life-cycle, and during the year. Most farming systems have a "hungry season" which often coincides with the time that school fees are due and life might be made much easier for many if the school year, or the fee-paying time were changed. Poorer people tend to borrow to buy food (at high prices). Then they are forced to sell their harvest when prices are low in order to repay the debt.

Rural differentiation: As the above example shows one man's poverty may be another one's wealth. The rich can become richer by selling their harvest when prices are high, and by charging interest on loans made to the poor. Some recent case studies (cited in Appleton et al, 1995) argue that rural differentiation is increasing, and that one aspect involves distressed sale of land and labour by the poor to the rich.

Inter-generational transmission of deprivations/advantage: Another question for investigation is the extent to which, and the processes by which, advantages and deprivations are transmitted inter-generationally. From this analysis points of intervention should be considered (eg an inheritance tax, scholarships for poor able children, basic health education for all).

The chances of living in poverty in Uganda depend very much where you were born (see Appendix 3 and the Map at the front). The different farming systems provide different opportunities and constraints, and are affected differently by shocks and policies. AIDS, malaria, tsetse flies, proneness to drought and physical insecurity, access to roads, schools, health services, markets, towns, electricity, water, and fuel depend on physical location.

Climatic and natural resources are unevenly distributed between the fertile and rain-sufficient lake crescent, through the drier surrounding agro-pastoralist zone, to the arid north-east. Five important farming systems can be isolated: the Central and Western banana and coffee system; the Nile farming system; the Northern farming system; mountain farming systems ; and pastoralism. The three main geographic regions of Uganda traditionally contained quite different societies - wealth was concentrated in the densely populated centre; in the western agro-pastoralist zone pastoralist elites generally dominated the agriculturalist majority. In the arid north pastoralism prevails. In addition there are several small specialized ecosystems: the islands in Lake Victoria and the remote mountainous or wetland areas. The recent discovery by the North Koreans of large deposits of minerals (gold, diamonds, marble and titanium) in the North may help to redress the regional balance in the future. At times regional disparity is politically motivated as it is with Kiboga district and the East Samya rural area or due to over-concentration of resources in certain districts in the Buganda district and, today, the Western region.

Historically social relations and distributions have varied between ethnic groups. Some societies were more unequal than others. Colonialism produced a peasant farmer commodity economy which suppressed the growth of an indigenous economy and created conditions for wealth, poverty, exploitation, and regional inequality. The fertile area became the central producing area while the north and Rwanda became labour-supplying areas and the isolated areas were further excluded from development. The centre became rich with pockets of poverty, the north became poor with pockets of wealth, while the north east remained in poverty (ActionAid Uganda, 1994).

Broad soil classifications by district show that the Central region is most favoured, while the Northern is least favoured:

The availability of agricultural land, while not limiting growth for the country as a whole, is a constraint in certain high-potential, high-population regions in the Southwest and Northeast. Landlessness and outmigration can be found in Kisoro, Kabale, Mbale, Pallisa and Tarrier. "The low carrying capacity of the land in areas like Kabale, Rukungiri, Masaka, Mbale, and Nebbi and Busoga districts has led to a growing number of unemployed young people in the countryside." (ActionAid Uganda L, 1994 p5). Historically there has been migration out of Kigezi (Kisoro and Kabale), West Nile (Arua and Nebbi) Bukedi (Tarrier), and Bugisu (Mbale).

By all the most crucial development parameters which include education, health, utilities, employment, and industrial development, the North compares less favourably with the south. Medical personnel in Gulu and Kitgum estimated malnutrition to be at least 60% and acute malnutrition in Gulu, Kitgum and Karamoja to be 30%. General deterioration in past decades, including the war that started in 1986, have had a more devastating effect there.

While regional poverty depends in part on physical location, environment and history, it also has political causes and consequences which will be discussed in greater detail in section 5.4. The civil war and ensuing insecurity had a larger impact on the North. Some areas have been particularly affected by war, particularly Mutukula in Rakai district, Arua, Gulu, Luwero, Apac, Kitgum, Soroti, Kumi and Pallisa. In these districts cotton ginneries, store houses, co-operative buildings, schools and health centres were damaged and looted in the wars from 1979 to 1990. Institutional infrastructure was also damaged including supply lines for agriculture inputs, markets, extension services, and bank credit facilities. People died and migrated eg people from Soroti to Kamuli. Arua experienced active war from 1979 to 1981 affecting 300,000 people. In Gulu, war broke out in 1986 and reached its climax between 1987 and 8; the most affected counties were Aswa and Nuwoya (ActionAid, 1994). In 1982 approx 95% of population of Moyo district had fled to nearby countries (UNICEF, 1994, p9 ).

There is still insecurity in some parts of the country disrupting the productivity cycle, affecting schooling and causing malnutrition. The worst hit parts are affected by cattle raiding and the Kony rebellion which is receiving support from Southern Sudan. There are also constant rumours of undisciplined behaviour by NRA soldiers in the north. One unpleasant aspect of rebel activity involves mutilation and the rape of girls. People have fled to military camps to escape the Karamajong cattle rustlers who are involved in looting and violence. Displaced families are living in dire poverty in Lira. The main conflicts are between two groups the Bokora and the Matheniko. There are refugees in Kumi, Soroti and Karamoja.

Regional poverty is also affected by the differential distribution of HIV/ AIDS with the effects of the epidemic most clearly visible in areas such as Rakai and Masaka which have been affected by the epidemic since the early 1980s. "AIDS is a district problem - because we are losing our future learned sons and daughters who would be the leaders of tomorrow" (Masindi interview - UNICEF Situation Analysis p66). Most people who fall sick with AIDS in Kampala are sent back to the rural areas. In urban areas with a higher seroprevalence there has been a noticeable loss of educated and skilled workers. In South West Uganda patterns of agriculture are already being affected. Families who have lost 1 or more to AIDS may shift from labour intensive cash crops (eg rice, millet, tomatoes) to less labour intensive, often less nutritious crops such as cassava and maize (Barnet and Blaikie) some fields in Rakai and Masaka are lying fallow after the deaths of those who tended them (UAC1993).

Appendix 3 contains a more detailed spatial description of the economy. Here it is important to point out the extent to which economic resources and capacity are unequally distributed; large areas are suffering from high and increasing marginalisation (Brett, 1992). Spatial location is critical. Modern development is concentrated in the Kampala-Jinja complex. Rural enterprise depends on local demand for cheap products protected by poor access and narrow markets, or services related to agriculture. Modern economic activities are highly concentrated in urban centres in the south (Kampala, Jinja, Masaka, and Mbale). Only a few district centres have genuinely "urban" characteristics (Mbarara, Kabale, Fort Portal, Gulu, Arua, Lira, Seroti, Tarrier); the rest are hardly more than trading centres plus administration. Only a few urban centres provide professional services or a differentiated retail sector. There are a few telephones, a district hospital, hotel, a branch each of the Co-operative bank, and the Uganda Commercial Bank. Supplies of water, electricity, and fuel are unreliable and there is no access to libraries or technical information.

Even those district capitals with "urban" characteristics may not be very well-endowed. For example, Fort Portal, the capital of Kabarole, "bears the scars of a rural centre teetering on bankruptcy, civil strife and prolonged government neglect. Ranked among the country's least developed towns, Fort Portal's entire infrastructure was devastated during the two liberation wars of 1979 and 1985. The recent killer earthquake caused yet more misery." (New Vision, August 31,1994, supplement on Fort Portal's centenary).

Figure 12 provides a start for causal analysis and suggests that the range of policy possibilities is vast. Many of these are discussed elsewhere in the paper; in this section we only raise two questions: why is Uganda so poor? and how poverty relates to social stratification.

Uganda's poverty may be viewed along the five dimensions of deprivation described above. Physical features include the facts that it is land-locked and has some areas with poor soil and low rainfall; it has low GDP per head and is in debt; nutrition, health, literacy and education are also low. Uganda suffers from political, cultural and social deprivation in the international sphere and is vulnerable to shocks, dependency and exploitation. Many of the population have been emotionally-affected by the recent horrors of war and AIDS. Children born since 1979 in Uganda constitute nearly half of the country's population by 1994. They have grown up in a country experiencing one misfortune after another; many have personally experienced violence or lost a relative to HIV/AIDS (UNICEF, 1994, P11).

With regard to external features the prices of Uganda's traditional export crops - coffee, cotton and tea - are much lower in real terms than they were in the early 1970s (World Bank, Agriculture, 1993, pxi). Uganda has suffered two recent major adverse external shocks: periodic droughts in 1991 which raised food prices appreciably and a massive decline in the international price of coffee from US$2.8 per kg clean unit value in 1986 to US$0.90 in 1992. The west has also been affected by the events in Rwanda.

The debt burden has an impact on Uganda's capacity to invest in its own development. Uganda has an extremely well-articulated debt strategy which the Bank supports and has successfully eliminated commercial debt using the IDA Debt Reduction facility. The debt-service to export ratio is planned to fall from 125% in 1992 to 35% by 1993. However, some argue that more needs to be done and claim that the debt crisis is a direct cause of poverty, draining the country of resources needed for investment in health/education and undermining prospects for future and current economic recovery. "The 1993 PER estimated that 13.6% of total expenditure (including that part financed externally) would be required to cover interest payments and about as much again on amortisation of principal. This total is more than three times the wage and salary bill, and about seven times the Government's contribution to the development budget (which is admittedly heavily dependent on donor funding)." (Goetz et al, 1994, p30).

One suggestion made to us in Uganda was that some of the operations of the international development management system were actually a cause of poverty. Evidence cited included the way in which multilateral and bilateral loans are used to buy Western products (for example, import support and studies by foreign consultants) which are more expensive and less appropriate, have to be paid back, and do not generate much further activity within Uganda.

Internal causes of poverty include systematic patterns of discrimination - based on class, gender, race, age, and ethnicity and the legacies of war, regional neglect, and political opportunism which diverted local and national institutions towards elite interests, resulting in, among other things, bad economic policies.

Current sources of income in Uganda include: aid (via government, donors, and NGOs), corruption, government rent-seeking, formal sector business, informal sector business, large agriculture, small agriculture, religion, local politics, trade, banditry and theft, "sharing" and charity. To the extent that the elite, which derives most if its resources via politics, diverts aid intended for poverty reduction programmes then poverty and social stratification are closely linked. The wealthy also gain access to agricultural and industrial (informal and formal) surplus via oligopolistic practices and price-fixing, controlling local grain and credit markets so that the poor are selling grain after harvest the price is at its lowest and buying it back when prices are high, and using the tax and regulatory system to acquire rents. Some of them also use their political power to engage in " land-grabbing" in some areas.

Capitalism is not yet established in Uganda although a beginning has been made. It emerges through two processes: artisans and wage labourers slowly transform themselves or the power of state is used to hasten the transition by providing the new class with access to resources. Brett (1992) claims that both are happening in Uganda. After 1962 an African entrepreneurial class emerged but in conditions which stopped it from operating on the legal-rational principles presupposed by economic theory. The rural economy is dominated by the political and civil service elite who retain their urban links and invest some of their urban earnings in rural businesses. They maintain personal connections and obligations in their home village. A few small entrepreneurs have accumulated resources by identifying local opportunities, often in the black economy, working hard, and applying an instinctive organisational ability. They often take advantage of resources derived through links with the formal sector.

The new business elite consists of people who have moved into the formal sector through education and contacts, acquired resources through primitive accumulation and developed their local assets through family networks. The most successful local entrepreneurs can develop political connections and use them to acquire assets in Kampala and develop them locally. They combine activities usually thought of as separate in modern societies: urban/rural; state/market; personal/corporate (ibid).

Rural accumulation has been dominated by surplus transfer via the state. This involved high official salaries up to the 1970s, access to Asian property in 1973, exploitation of monopoly rents derived from official positions since 1962, and acquisition of land through the 1958 land legislation. As a result the political and administrative elite has wide interests in local transport facilities, hotels, shops, trade, small processing industries, large farms, and ranches. Much of this capital came from illegitimate sources particularly in the 1970s and 80s. As these businesses have succeeded owners have grown more reliant on business incomes than rents and this could produce a significant change in the nature of their position in the economic structure and attitude to the state.

While the development of capitalism may be necessary for growth and development in the long-run, if it is not controlled then inequality and poverty are likely to increase as a result of the scramble for resources. The problem about this is that there is currently not a class of entrepreneurs and a political class who might check the worst excesses of the other. Rather there is a more homogenous elite pursuing its self-interest in whatever arena appears profitable.

For sociologists poverty is a political matter as much as an economic one. It is embedded in unequal economic, political and social relations, and transmitted through the generations culturally. It is something experienced by people as they pass through life and involves a range of deprivations, processes and relationships, some of which are more amenable to possible policy interventions than others. A strategy for poverty reduction with any chance of success must be a long-term one. It requires integrated attacks on many fronts including the legal, political and cultural and particularly the organizational as well as the economic. Such a strategy cannot be developed as a technical matter - it is political and involves values, interests and public discussion which and must come from the Ugandan people themselves. A poverty strategy based solely on the economic model described above and imposed by the Bank as part of conditionality will not work. This style of interaction hinders the development of "ownership" and "accountability" since it is the Bank that owns the policies and the Bank to whom politicians and civil servants are (theoretically) accountable. Future research and policy advice relating to poverty in Uganda should be managed and done as much as possible by Ugandans as part of an ongoing process of data collection and policy and recommendation and evaluation. Given a Ugandan commitment to poverty reduction, both economic and sociological analysis can play vital roles in the development of efficient, equitable and implementable policies and research programmes geared towards poverty reduction. The sociological preference is for providing analysis and conclusions in a form which allows for choice: if you do this then this (intended) and this (unintended) might happen rather than advising one path.

The conceptual model outlined in Figures 5, 6, 7 and 8 can be used to analyse a number of social processes with implications for poverty, economic growth and development. Certain features of the developing system and social configurations in Uganda since Independence have resulted in the playing out of certain system contradictions and social competition and conflicts, one consequence of which has been the deepening of poverty. Here, we only have space to consider a few of the possibilities that occur from studying the model. These are institutions and rural productivity; relations between the formal and informal; coping strategies and investment; and why Uganda is so much poorer now than in the 1960s.

There are a number of institutional, cultural, and organizational factors which might affect the supply response to trade liberalisation (see Figure 13).

CAN YOU SCAN IN FIG 13?

Here the focus is on a economic meso action variable: middle men. "In the marketing of crops, large private traders have in many places occupied the monopsonistic positions left by co-operative societies using small scale traders (bicycle traders) as intermediaries. Marketing margins are high, extortionately in some remote areas. Intra-seasonal price variations are large and problems of storage, especially of maize, make it difficult for the majority to take advantage of seasonal prices." (Evans 1994)

This question concentrates attention on a cultural meso system variable: the gender division of labour. Women produce subsistence crops and do such things as the weeding of cash crops. They already work an estimated average of 15 hours a day. In the rural areas cash is controlled by men; in many areas even if the woman sells the goods she hands the cash to her husband (see Appendix 1).

Here the importance of time and the creation of expectations is raised. The chief concern of the peasant household is survival; if prices are unpredictable from year to year it is difficult for the farmer to make rational decisions based on market trends. In this environment the spreading of risks, including social investment, seems a more rational decision than responding to a price increase which may not apply to the next harvest. The extent to which increasing rural productivity depends on specialisation is a matter for further investigation.

A large proportion of rural farmers use the hoe and no fertilisers, improved seeds or pesticides. Some don't even have hoes. The agricultural extension service is underpaid, often does not have finance for vehicles or petrol, and is unequally distributed throughout the country. "The bureaucracies are unresponsive to field-level needs and priorities." (Uganda - Agriculture: World Bank 1993). Figure 13 suggests a cultural class between the training agricultural extension officers receive and the realities of rural production. For example observers report gender bias in the National Agricultural research Organization. The question then arises can improvements in technology be disseminated to the right people outside the public sector, and if so how?

At the bottom of Figure 14 a number of coping strategies are identified. Obtaining food, cash etc in the circumstances in Uganda today is not just a case of entering the labour, land or product markets. It involves negotiating a whole series of relationships and possibilities in the political, cultural and social fields as well as the economic. During Uganda's crisis, which lasted from the early 1970s to the mid 1980's people developed a range of ways coping, the most important of which were withdrawal from the formal economy, privatisation of publicly-provided facilities, diversification of activities, and "sharing". These are discussed briefly below since they have implications for policies aimed at poverty reduction. Appendix 1 contains a few examples of coping strategies adopted by some of the particular vulnerable groups identified earlier.

CAN YOU SCAN IN FIG 14?

The urban informal sector in Uganda is thriving, as elsewhere in Africa, but not much is known about the extent to which activity is concealed from official statistics. Smuggling is still a feature of economic life, although it is not so widespread as earlier. Brett has recently described Uganda's rural areas as deprived and often dangerous places; what goes on there is at best semilegal and involves considerable concealment to avoid tax, customs, policeman, and the moral censure of church, elders, officials. Brett suggests that MacGaffey's findings in Zaire (MacGaffey, 1987) closely parallel his own in Uganda: from the official figures on wages, prices and the decline of real income it would seem impossible for the majority of people to stay alive in Zaire. However some do very well and there is a highly organised system of income-generating activities. While these economies miss a developed capitalist component they are highly differentiated and involve a dense mass of smallscale activities not only for the domestic market but for national and export markets as well.

In the local exchange economy almost everyone sells some agricultural output and at least some members of every family will be involved in agricultural processing and off-farm activities. In the markets people sell grain, fruit, vegetables, fish, and meat; the food is often processed. Services provided include transport, bicycle-mending, cutting hair, and making clothes. Products made and sold locally include furniture, pottery, shoes, woven mats and baskets, sisal ropes, grinding stones, paraffin lights, charcoal stoves, spear heads, axes, and metal hinges. Building materials including bricks, tiles, timber, bamboo poles, thatching grass, and stones are sold on the roadside or produced to order. Personal services are provided by traditional healers (spirit mediums, herbalists, daktari, and birth attendants) and are the main source of medical treatment for most. Women and children brew beer and spirits, musicians play, travelling players perform, and sex is sold. Brett suggests that the income generated by these activities must be very large. (Brett, 1992)

This was a strategy adopted by almost everyone in Uganda at some time since Amin took power. There may be political reasons for this as well as economic since in the past Governments (late 1960s to 1986) would issue decrees affecting particular groups, such as shopkeepers. This meant that people needed a range of activities in order to cope. Thus urbanites have taken up urban-farming, farmers have gone into trade and diversified their crops, civil servants have formed NGOs, driven taxis, and gone into crop trade, while some of their wives have opened kiosks. The effect of this diversification has been to undermine the process started under colonialism of an increasing division of labour, as occurred in modern societies. When people do a bit of everything they do not develop an occupational identity - such as civil servant, landlord, wage labourer etc, and their activities cannot easily be measured and analysed using concepts developed for a more definitive situation.

Privatisation of publicly-provided facilities has been adopted as a strategy by civil servants including health workers and teachers who have been unable to live on the routine government salaries and benefits which have often been paid very late, if at all. They have engaged in a variety of income-producing strategies involving the use of government facilities as private-for-profit institutions where the material and financial inputs from government serve as a subsidy. Doctors have introduced informal charging and the sale of government or donor provided drugs. Teachers provide private tuition after school and in some cases use the labour of pupils to grow food or other crops. Staff at Makerere University have rented out part or all of assigned staff quarters, including their garages. People have volunteered for large numbers of short courses and training for the per diems. "With two decades of history, the structures which health workers have developed as methods of survival have had the opportunity to become entrenched and will be difficult to eradicate or modify." (UNICEF, 1994 p40).

5.2.4 Negotiation

Recent historical and anthropological literature has been marked by a "growing tendency to portray African cultures and institutions as fluid and ambiguous, subject to multiple interpretations and frequent redefinition in the course of daily practice" (Berry, 1993, p4). Negotiability is a key feature of economic and social relationships, one example of which is the way in which government officials implement rules and regulations in what has been described as an arbitrary and capricious fashion (Lancaster, quoted in Berry: p13). "In economic (as in other forms of human activity) people are continually engaged in efforts both to influence and to understand the circumstances under which they produce, invest, exchange, and consume goods and services. Producers, for example, seek to bring power to bear on acquiring and using the means of production" (ibid).

Berry argues that during the colonial period African farmers in many places became increasingly involved in market exchange but social networks still remained an important channel for access to productive resources. At the same time colonialists tried to incorporate customary systems of law and authority and reorganize traditional polities according to Western conceptions of administrative stability and efficiency. This often led to prolonged debate about the legitimate interpretation of customary rules and the demarcation of traditional hierarchies and political jurisdictions. As agricultural production for the market increases so too do resulting struggles over access to land, labour, and capital. This conflict fed into and intensified ongoing debates about meanings of customary rules and structures of authority and these were rarely resolved in a definitive manner. The ability of Africans to exercise influence in the colonial order was linked to participation in these debates about local institutions and the ability to participate in the debates depended on social identity and status as well as command over material wealth. This meant that people were encouraged to make social investments, and this reinforced the importance of social relations as mechanisms for access to productive resources.

Social relations continued to provide significant channels for negotiating access to resources after independence; most government and political movements mobilized popular support by appealing to sectional identities; social identity and status became important as channels of access to the state as well as local resources. Sectional politics and socioeconomic diversification in the face of uncertainty have further reinforced the proliferation of social networks and their continued importance as channels of access to productive resources and economic opportunities. Today in most African societies access to land continues to be mediated through membership in various social networks, although Berry argues that access to labour has become increasingly individualized.

Berry suggests that if rules, transactions and values are ambiguous and negotiable and access to resources and opportunities depends on the ability to negotiate then people may be more interested in keeping their options open than cutting them off, and in strengthening their ability to participate in and influence negotiations rather than acquiring exclusive control over resources and severing connections which are not immediately profitable. The literature on African rural economic life is filled with examples of farmers who prefer not to register their land rights. employers who don't dismiss redundant or unproductive workers, labourers who work without pay though hiring-out is an option, lenders who do not charge interest, and borrowers who pay back more than they owe (p 14) In many societies a person's status and influence depends directly on the ability to mobilize a following. Ambiguity in institutions has given rise to both inclusive and exclusive strategies of social mobilization and resource access: people invest in the means of negotiation as well as the means of production; agricultural surplus is used in part to establish and strengthen social identities and relationships through which people enter or influence negotiations over access and control of resources.

In the recent crisis Ugandans have continued to make large social investments to gain access to resources and to spread risks. One way to spread risks is to invest in another individual in the expectation that if you are in need he or she will help you. So you may invest in the school fees of a relative or the child of a friend who you know to be clever. By marrying outside the village and paying or receiving bride wealth you may create obligations that can be called in time of need. You may decide to have children as a hedge against future insecurity. Children are wealth; large numbers of children strengthen and build the clan and provide an expanding labour force. Parents see children as an investment in the future (UNICEF, 1994 p52). Another way of risk-spreading is to lend to a friend or relative in need without interest or time limit to create an obligation which you might or might not claim back in the future depending on the circumstances of both parties. A related coping strategy may be described as client-seeking behaviour - for example, you lend a piece of land to someone on the understanding that he is available for service when you need him. The converse of this is patron-seeking behaviour

People also invest in social groups. They invest material resources in feasts of various kinds which create social obligations and, to the extent that they can, they make status seeking expenditure. Time is also invested in socializing and developing networks of support and friendship groups. The wider and more dispersed one's network is the more likely it is that a crisis will be survived satisfactorily.

When resources and time are scarce it is often a choice between productive investment or social investment; it is possible that one explanation for the very low savings and investment rates in Uganda is that the resources are being invested in other people.

A recent article by Bangura (1994) has suggested that there are two key contradictions which might explain why the civil service and other public sector organisations have been largely ineffective in the crisis economies of Africa. These are the growing contradiction between the interests of bureaucratic actors and the goals they are supposed to uphold; the other is the contradiction between the formal rules and regulations which are meant to govern bureaucracies and the emergent institutions of the wider society. These contradictions are depicted in Figure 15. As we have shown in section 5.2 between 1971 and 1986 the Uganda economy, society, polity and culture grew increasingly informalised. Declining living standards and insecurity led to crises in social cohesion and compromise, institutional socialization and loyalties, clashes of public and primordial values, and the failure of political authority to enforce rules and regulations. The question now is whether and how attempts should be made to drive back the informal with a "top-down" assault (and if so who is to do it since there are few Ugandans who are not embedded in the emergent informal structures) or whether imaginative use should be made of these indigenous structures.

Uganda is very poor with a GDP per capita estimated to be $170. The World Bank estimates she is the 22nd poorest country in the world and the 13th poorest in Africa. If we are looking to explain why Uganda is currently so poor it is necessary to go back to the 1960s. In 1962, when Uganda became independent she had "one of the most vigorous and promising economies in sub-Saharan Africa" (GOOP, 1993).

Uganda was self-sufficient in food and the agricultural sector earned a large amount of foreign exchange, Textiles and copper were exported and there was a current account surplus. Infrastructure and health services were some of the best in sub-Saharan Africa. Immunization programmes reached as much as 70% of the population. In 1970 Uganda had the fourth highest GAP per capita in sub-Saharan Africa.

Following the 1900 Agreement the British established a system of administration in which districts demarcated along ethnic lines became the basic political and administrative units (Barongo, 1989). The areas were given considerable local autonomy, particularly the Kingdoms, and most particularly the Buganda. Following the 1949 Local Government Ordinance the British encouraged educated Ugandans to work in their home districts, leaving the central government to British officers. Local government became the focus for African participation, identity and loyalty. Local decisionmaking power was decentralised and local governments controlled critical resources, including employment.

CAN YOU SCAN IN FIG 15?

The 1962 Independence Constitution was designed to minimise ethnic conflict by granting greater autonomy to kingdoms and districts, establishing a strong system of local government, with elected councils for each district. It only lasted four years. Milton Obote's UPC party, in collaboration with the Protestant Kabaka Yekka party, won a majority in the first general election. However, the central government began encroaching on the powers of the federated states, the relationship grew strained and local units entered the central political arena to assert themselves. The central government reacted and centralized more. The Buganda crisis of 1966 led to the end of the Constitution and the 1967 Constitution abolished the kingdoms. In 1969 following presentation of his socialist package Obote was shot at and this was used as the excuse to establish a one-party state. The area for political competition grew smaller and smaller. To retain power Obote appointed his most trusted men to important positions. They happened to be Langi or Acholi and the other ethnic groups began to feel politically dominated. One of the charges made against Obote after 1971 military coup led by Amin was of ethnic domination. However, Amin proceeded to do the same: he eliminated the Langi and Acholi from army (by massacring a large number) and then established a hegemony of economic and political power based on a coalition of ethnic groups from West Nile and southern Sudan. He declared "Economic War", deported the Ugandan Asians and redistributed their businesses and land. Economic mismanagement and political terror was accompanied by the departure of many educated Ugandans, and an embargo by the West on trade and aid. Physical, economic and social infrastructure decayed.

Amin fell in 1979 and after an interim Obote returned in 1980. Following rigged elections Obote once again established a government propped up by the Langi and Acholi in the Uganda National Liberation Army. The current President, Museveni organised the National Resistance Movement and the NRA (guerilla army) both of which came to be dominated by southern Bantu - particularly Banyankole and Baganda. The civil war began in February 1981, Obote was ousted by the Okellos (Acholi) in 1985 but they then lost to the NRA in January 1986.

This description of progress of Uganda towards poverty involves both system "contradictions" and social action between individuals and interest groups. The major system contradiction is one between ethnic pluralism (meso level - cultural) and political centralization (macro level - political). This was enacted out in the political and military arena by mega actors and meso group actors (Figure 16)

Ethnic pluralism involves multiple centres of interests, identities, loyalties, patterns of action and power. Political centralization involves the creation of decision-making organizations and institutions run from the centre. When access to resources depends on political access to the centre rather than economic productivity, and when those resources are scarce, economic competition is replaced by political competition which usually involves use of the means of violence, surveillance, and organisation. If political competition is defined in ethnic terms then questions of language, identity, and loyalty arise. Ethnicity becomes an important factor not only in politics, but in the armed forces, the bureaucracy, and in business. Political struggle involves not only a struggle to occupy office, but also a changing of the institutional rules to benefit the incumbent group.

One way to diminish the contradiction between ethnic pluralism and central government is to decentralise, and this is what is currently happening in Uganda under the Museveni government (see Section 2.3.3 for further discussion).

Our model offers other possible conflicts and contradictions for exploration including, for example, conflicts within the economic system domain, conflicts between the economic and political systems, and between actors trying to succeed in the different arenas. Uganda, like most African countries must be regarded as "unintegrated" both in terms of systems, and in terms of social actors. To analyse these structures as if they were comparable to those of modern societies is probably a mistake.

In this section we find ourselves making recommendations. However, we are aware that they are personal judgments and that they have the status of advice that we think should be considered even if it is rejected, rather than instructions that must be followed.

6.1 Contributing to the economists' debate

One of our conclusions is that obstacles to economic growth and implementing pro-poverty policies involve a complex, inter-related mixture of institutions, culture and politics. Above all economic lives in Uganda are chiefly organized via fluid social networks, rather than formal organizations or corporate groups and these social networks also operate within organizations which have been formally designed, and have the effect of undermining their formal goals. Research is needed as to how these networks actually work in Uganda and how they might be used to improve productivity and reduce poverty. Economists have a lot to offer in developing new models of exchange processes in African economies (see for example the literature on share-cropping) but they will have to work in a multi-disciplinary way in order to establish what it is they ought to be modelling and to contextualise their conclusions before proffering policy advice.

A second conclusion is that the statistics that economists use could be improved considerably with assistance from anthropologists and sociologists. How this might be done in the Ugandan context is spelt out in some detail in section 6.3.

Finally there would be considerable merit in developing links between macro, meso and micro research and applying them in future empirical research on Ugandan economies. At the meso level possibilities developed within the New Institutional Economics might be considered.

There is currently a large gap between what the "World Bank" says should be done in terms of measuring and analysing poverty (see for example, World Bank, 1990) and what actually the "World Bank" has done in this field in Uganda. The 1992/3 Integrated Household Survey which can not be used confidently in terms of its measure of consumption is rumoured to have cost the Bank $2million and there has been no serious analysis of it by Bank staff.

In the long-run Ugandans have to decide whether and how to make poverty reduction an integral part of their long-term economic strategy. Should they follow a "Scandinavian" development path prioritising equity and low inequality. Would it be quicker and more appropriate to go for growth and "trickle-down"? Or is there a middle path? Without debate there is not likely to be a poverty reduction strategy which works, for two reasons: it is not in the immediate interests of an elite which has still to develop a strong sense of the national public interest; and without debate and the adoption of a long-term strategy they will face the alternative of regular short-term shifts in policies according to the latest fashion in donor circles.

Ownership of the poverty agenda, long-term planning and programming, and sustainability are not likely unless people, and particularly the elites, are aware of the dimensions of the problem, have considered and discussed the many causes involved, and have themselves developed a programme and organizational structures for monitoring poverty and implementing pro-poor policies. The first thing required is a political debate conducted at all levels of society about priorities in development including poverty and welfare. Acceptable levels of wealth should also be discussed. This debate should be initiated at the highest level and encouraged to take place in the media, by politicians at national and all five district levels, among academics and donors, and generally among the population.

Macroeconomic policies: The government needs to maintain macro-economic stability if there is to be growth. However, the structural adjustment agenda does not belong to the government or the people of Uganda, and it is not clear how poverty-oriented it actually is. It is negotiated secretly between the World Bank and a small coterie of government technocrats, and is not explained to the people who it is affecting. The government should conduct an ongoing public debate about economic policies with politicians, non-governmental economists and other participants in civil society.

Social sector spending should be targeted to the poorest, and agricultural extension advice and research should be aimed at all levels of farmers (including women). The government should consider developing a parallel domestic extension and research service which could develop domestic technology in conjunction with the people who use it. This is particularly relevant for saving women's time (in grinding, fetching water and fuel, washing, and cooking). Mobile health workers could also save people a lot of time. Service provision is not just a matter of spending; services need planning to fit need and demand, appropriate training, monitoring and adaptation. The content of services is also important, for example the curriculum and the health advice. These are areas where public debate and participation would help to produce more meaningful formal structures and policies. Inventiveness and imagination in producing local solutions should be encouraged and successful ones widely disseminated.

Policy implementation: Problems of implementation involve institutions, organizations, politics, and culture. These extra economic factors should became a focus of attention in terms both of research and programmes. While many of the issues raised here have been, or are being approached in piecemeal fashion, it is unlikely that major transformations will occur without systematic analysis and imaginative and integrated projects and programmes, many of which should be initiated and supported outside the Government sector.

There are two major problems with policy delivery from the centre: capacity and corruption. These are currently being addressed by the civil service reform ( described above) but it is not clear, in the light of the discussion about the interaction between informal cultures and formal structures whether this reform will be sustainable. Capacity can be improved by on-the job training and the development of rules and cultures in conformity with planned goals. Reducing corruption involves changing the incentive structure: this means exploring the relationship between aid and corrupt practices, monitoring and sanctions, improving accounting practices, providing alternative productive incentives.

` Developing capitalism: The development of capitalism depends on creating an effective system of economic management; this means creating and maintaining really competitive markets and ethical business practices. The new entrepreneurial class is very weak and there are strong vested interests in monopoly practices. Effective policies are needed to eliminate the "parasites" and create conditions for rural farmers and small enterprises to raise levels of organisation, technology, and productivity. The changes needed are relatively easy to identify but putting them into effect is more difficult in face of opposition from vested interests and the weakness of the Ugandan state. There is a need to develop a legitimate business class which maintains high ethical standards in its commercial practices and observance of the law and a political and administrative class which does not behave as it did in the past. This involves developing mechanisms of political accountability and professional associations. These exist now but are weak and need more support. Because of still-existing regulations small businesses cannot obey the law and operate profitably. This criminalises the business class and breeds disrespect for the law (Brett, 1992).

The overall impact of state policy and intervention has been negative: many onerous and restrictive taxes and control are still in force because of bureaucratic vested interests or inefficiency. The other side of this is that many businesses have been financed as a result and most businesses manage to evade them. However there are heavy costs which make it difficult to make the transition from informal to formal levels of scale and organisation. A new capitalist class has emerged by appropriating resources from the state and exploiting the monopoly profits derived from irrational controls and scarcity. These are processes that occurred in the development of virtually every advanced capitalist society and are taking place all over Africa Brett, 1992).

Cultural adjustment: The NRM government "is committed to revolutionary reform of the psyche of Ugandans in order to build a new and viable basis for the nation" (Evans, 1994). Perhaps the most important aspect of this is changing gender attitudes and behaviour. It is impossible to disagree with Ali Mazrui when he says that almost all governments on the African continent have tried to steer clear of the sensitive issue of Africa's sexual division of labour (1992, p358). Only the state is strong enough to make the necessary cultural readjustment on the gender question. Only the state can enable the market to benefit from fuller female participation in the economy; "a true free market ought to be an androgenysed market" (ibid, p359). However, the road to creating female entrepreneurs is not likely to be through privatisation, but through energetic state action. The Ugandan government must follow-up on the implementation of pro-gender policies. This means integrating gender differences into all development and poverty programmes. Policymakers in all ministries should get information by gender and plan according to women's needs as well as men's. Findings of research and needs assessments on women should be incorporated into ministerial plans and strategies. Some women in the educated class have embarked on advocacy for women's issues through the formation of networking groups and use of the press and other public fora. These developments should be welcomed and encouraged by the government.

The President has provided a strong lead in reducing gender bias. The appointment of a woman vice-president is just one instance. However, there are still problems of implementation related to the various initiatives that have been taken. Thus, while there is a women's representative on RC committees right up to district level these offices have not effectively succeeded in influencing the decisionmakers because the women on these committees are often ridiculed and overruled by their male counterparts and some have no training (ActionAid Uganda, 1994 p16). The parallel system of women's committees operating at all five levels will provide opportunities for many women to participate in political activity if they can find the time but there is a danger of marginalisation. An alternative or additional proposal might be for their to be 2 women's representatives at each RC level which would make it easier for the women involved to deal with ridicule.

The women's representatives on the NRC are an important source of information about the diversity of women's conditions and this source should be systematically tapped. The preferential admission of girls to university by topping up their scores by 1.5 extra points is a pioneering example of positive discrimination which should be followed up with similar preferential policies at all other levels of education. It was suggested to us that the Government should re-introduce a school fees subsidy for girls. The post-graduate Womens' Studies course recently introduced at the University should be regarded as the beginning of a larger programme of research and teaching on Women's Studies. The programme of FIDA(U), Uganda's association of women lawyers should be expanded to cover as many districts as possible. The association strives to take the law to lay people so as to create awareness of rights and obligations of individuals; they offer legal advice and counselling for families. It assists women and children, especially widows and orphans, to attain effective protection under the law. They have been holding legal aid clinics in Kampala and Mbale for women who are unable to afford private firms. They also run legal education programmes and translate the law into local languages. Other women's organisations should be supported and helped, maybe through twinning arrangements with women's groups in the West. There are numerous organisations including the Uganda Women Finance and Credit Trust which was founded by women in banking, business, and law. Trade unions have a women's wing. Safe Motherhood co-ordinates women's NGOs involved in health related programmes; it has trained pregnancy monitors in 8 districts. There is also a Ugandan Association of University Women. The UNDP has financed the Uganda Women's Umbrella project in the Ministry of Women in Development, Youth and Culture to promote gender sensitivity among decisionmakers and policymakers at various levels in government . DANIDA also assists the MWIDYC.

However, on the whole women continue to be the recipients of development programmes not active participants in them. The marginalisation of women is likely to be increased by the enlargement of scale of African economies, the internationalisation of African economies, and the mechanisation of production. Increases in agricultural productivity depend on improved simple technology. Tractor schemes rarely work in Africa and the government should consider developing an oxen programme (and maybe donkeys). This will require dealing with the problems in Karimoja. The fact that men plough will reduce the hoeing burden on women, but it will increase the load for weeding, harvesting and processing, some of which could be relieved by simultaneously providing access to pesticides and mills. It might be worth considering teaching women in female-headed households to plough: even raising the idea would stimulate discussion. Trying to make women active involves conscientisation including gender campaigns about women's rights. The place to start is the primary school where there should be gender roles guidance and counselling for both teachers and pupils of both sexes. Social mobilisation and networking at local, regional and national should be encouraged. Government should identify and monitor the enforcement of existing laws that favour women and repeal discriminatory laws. They should identify, strengthen and give a high profile to indigenous women's organisations. The land reform should give equal rights to men and women.

Designing appropriate "top-down" organizations: Ugandan formal organizational life is marked by rigid formal rules which delay implementation and can only be got around by applying informal practices. Emergent structures and norms are often designed more to meet the goals of participants than of the organization. It is important that an understanding of organisational options is developed in Uganda and rules, norms and cultures antithetical to efficiency and effectiveness are displaced. Organisations can make use of networks, hierarchies, markets, or group participation to achieve their goals and the suitability of structures varies according to the goal that is being pursued and the culture of the people involved in the organization. Brett has suggested (1992) that the group method works better for services benefitting from economies of scale to which people feel access should be open, while it does not work so well for income-generating activities. Collective action is much more effective when it is concerned with the provision of capital, training and information than in productive activity. Training in basic business skills and dissemination of collective research is best done collectively. Bureaucracies can mobilise professional and technical services and develop and monitor standards. Networks can be useful for generating trust, and diffusing risks and information. Exploration of emergent organizational practices and beliefs and how they might be adapted into larger structures would be useful. There is scope for some creative and imaginative interaction between new Western organizational theories and participative practices being encouraged particularly by NGOs. Ugandan bureaucratic practices have developed out of organizational designs suitable for the West in the 1950s and 60s and it is not surprising that they have been subverted. Redesign with assistance from participants in relation to organizational goals they have explored thoroughly might lead to better results.

If the RC system is to work certain anomalies will have to be eliminated. First there has been no demarcation of tasks between officials and politicians associated with clear lines of authority and responsibility. Four lines of authority have existed since 1986: the RCs, the District Administration, line Ministries and the District Administrator; the RCs answer to the electorate for the performance of officials who they have no power to recruit, deploy, control and discipline. Decentralisation is expected to give much greater power to elected councils. There is also a need to delineate areas of administrative responsibility and allocate them to particular levels of the system. RCs are responsible for local roads, primary education, health centres and dispensaries, boreholes and springs, petty crime, domestic and land disputes, but Brett (1992) suggests the responsibilities are not allocated properly. He also thinks the number of levels should be cut; RC4s do virtually nothing in the rural areas and it is "hard to make a case for RC2s" .

Following decentralization there should not be unacceptable inequalities between districts. Due to economic inequalities and political instability there are great differences in income-generating capacities between the wealthiest and poorest districts: Mukono recently generated Sh 650m while Kumi only raised Sh 50m. Between 1962 and 1991, due to factional conflicts and ethnic rivalries, the number of districts has increased from 16 to 39. Many of the new districts do not have adequate administrative infrastructures and there are a few marginal administrations which are incapable of providing effective services. The central government should monitor the affairs of the districts and consider making equalisation grants.

It is probably unrealistic to expect the RCs to alter the balance of power between rich and poor in local communities because the higher level councils are composed of the traditional elites and women are virtually excluded because they have no time. There is no distinct working class in villages to provide the basis for a party of the have-nots and political competition takes place between individuals mobilising patron-client networks. The political elite tries to increase the resources coming into District and ensure that they plus their families get favoured access to what is available. The democratic process cannot give the poor control in these circumstances but should help them make those in power actively promote the interests of the locality and behave with some degree of honesty and competence. There is a need for constructive approaches on the part of government, donors, and NGOs towards ensuring that the decentralisation process delivers on its promise of orienting public spending towards the interests of the poor. The next PER should explore the means by which local-level resource scarcity can be mitigated and effective auditing and accountability instituted.

All government policies and programmes should be regularly audited for their effects on distribution and poverty. Poverty programmes and research should be centralised in terms of organisational structures, and prioritising.

6.2.4 Policies for political stability

As section 5.4 demonstrates political instability is a major cause of poverty and has partly resulted from competition between ethnic groups for control of resources concentrated at the centre. Decentralization of resources and management should reduce this competition and the salience of ethnic politics, if it is a real devolution. However, as discussed above, the danger with decentralization is that regional inequity will increase; some form of redistribution from rich to poor areas must be institutionalized. This is particularly important in the provision of structures, services and regulation relevant to development. In poor rural areas there is not only an infant industry problem but also an infant region problem (Brett, 1992). Assistance is needed to create the array of public and private investments required to retain people and attract them from outside; only the state can reduce competitive inequalities between poor rural areas and their developed counterparts. For while political instability is a cause of poverty, poverty, particularly unequally distributed poverty, can be a cause of political instability.

On occasions there is a conflict between the economic objectives of efficiency and equity and the political objectives of stability and fairness. For example the World Bank Agriculture report says "The role of Government in developing strategically placed rural infrastructure cannot be overemphasised." pxxiii. However, what is strategically placed in terms of economic efficiency may not be strategically placed in political terms.

One problem related to dealing with corruption is that it has been seen as an important factor in the maintenance of political stability. However, economic growth and development are not usually associated with a political structure based on distribution via patron-client relations, and donors will not continue indefinitely to provide the resources which are currently being diverted from development investment and social sector spending and channelled down these patron-client networks. A clear message that corruption does not pay must be given to politicians, the civil service and the public and corruption must be made not to pay.

The African crisis is very complex and has affected all strata and sectors of society; its dynamics have been shaped by inter-locking economic, social, political and cultural factors inter-acting through time. For various reasons the IMF and the Bank have refused to explore the full dimensions of the crisis, but instead have tried to tackle it using a simple market-oriented policy framework (Bangura, 1994). During the 1980s African governments lost effective control of economic policy-making to these international financial institutions who have tended to make blanket recommendations to all adjusting countries, rather than helping individual governments to work out a strategy tailormade for the total circumstances of the particular country.

With regard to poverty assessment and policy recommendations in Uganda the Bank approach is still very "technical" (compared for example with the approach in Zambia) and this allows the Government not to take it seriously as a policy matter. The assumptions that structural adjustment (as practised in Ugandan conditions) will lead to growth will lead to poverty reduction are still assumptions. Some economists expect that they will; most sociologists and political scientists expect that structural adjustment might lead to growth for some, but will probably lead to more and/or deeper poverty for others. Policies aimed at reducing poverty need to identify which groups of people are poor, vulnerable, getting poorer, and what their circumstances and strategies actually are so that policy levers can be identified. While, at some higher level, the Bank is claiming that poverty reduction is its major goal, this does not seem to be the case in Uganda as is evidenced by "Growing out of Poverty", the current Country Economic Memorandum, the low profile for poverty in recent Public Expenditure Reviews and in the cash accounting priorities, the difficulty of seeing how Bank loans relate to poverty reduction, and the unavailability of an analysis of Bank-financed projects in terms of their goals, monitoring and evaluation.

Furthermore the elites of Uganda, who control the major resources, have not shown much interest in poverty reduction and there are no powerful voices speaking out at the national level on their behalf (in contrast, say with Victorian Britain), although there is some concern at lower levels. The "de-politicization" of the question of poverty provided by the Bank approach means that elites can continue to accumulate resources by appropriating rents from their involvement in politics/administration, diverting government funds (many acquired as aid) and appropriating and controlling land, labour etc in the rural areas, without this being raised as an issue.

The idea that governments bear some responsibilities towards its citizens should be encouraged to take root in African countries. The first obligation a government has is the maintenance of order within its boundaries, the second to ensure that citizenship rights are protected and exercised. There is a strand of opinion in the aid world that recommends the bypassing of inadequate governments in the provision of essential services, such as health and education which are basic citizenship rights, and the channelling of resources and services through privatization and the NGO sector. Experience throughout the world confirms that the privatization of services such as health and education increases inequality and poverty (measured in terms of access to and quality of these services). Similarly in developing countries the provision of services through NGOs, given their size and the control they have over their agendas, results in a patchy and inadequate provision. The Ugandan government does not have the resources to provide cheap universal basic health and education services but this should be a long-term goal and, in any case, the government should be in charge of the design and regulation of these services.

Macroeconomic policies: The international financial institutions should consider the contradiction (one of a number) between their practice of "negotiating" macroeconomic policy with African governments and their call for "ownership" of poverty reduction programmes.

Social sectors: Similarly donors should not allow their need to "disburse" loans or grants to interfere with internal processes of policy development and debate as has happened in the field of education in Uganda, and probably elsewhere. It is better to give or lend no money than to disrupt internally generated policy and project preparation.

Policy implementation: Donors are much better at drawing up blueprints than implementing programmes and projects. The latter requires intimate knowledge of a range of structures, relationships, values, beliefs etc and a long-term commitment. Donors should consider the value of personnel based in foreign capitals a long way away who know little about the countries they are advising and "researching".

Cultural and political adjustment: Donors have, in the past, steered clear of getting involved in recommendations that governments should act to try to change the values, beliefs and attitudes of their citizens. Caught between the economist's commitment to accepting people's preferences as given and the post-modern acceptance of the notion that it is impossible and unethical to make judgments as to the superiority of some values compared with others, they have tended to avoid the question altogether. The international financial institutions have the added problem that they are not meant to interfere in the internal affairs of the countries they lend to, especially the politics. Given the role that aid is already playing in the political economies of sub-Saharan African countries this is a contradiction that will have to be resolved soon: it might involve a re-writing of the Constitutions of the Bretton Woods organisations, or the creation of a new organisation designed to deal with the problems of the 2000s. In the interim there is no reason why the IFIs should not try to understand and study the political, cultural and social aspects of African economies, even if they do not develop policy advice in relation to them. In fact the promulgation of "multi-party democracy" must be regarded as beyond the remit of these organisations. It is also being recommended in a blanket and unthought-out way with little attention and research (which should be done by political scientists) to appropriate structures of government and democracy which will vary between countries.

Donors might set a better example in their own gender policies and practices.

Finally donors should consider the way in which they work in developing countries. For example, the Ugandan elite has high aspirations in terms of the lifestyle they are seeking and it easy to believe that the Kampala diplomatic and aid community is a major reference group. A more modest life style on the part of resident donor staff might make it easier for them to convince the Ugandan elite that it should be actively concerned about poverty and sacrifice some wealth and wealth-seeking activities to achieve this.

Organizational design: There are exciting changes in the design of organisational structures taking place in both the public and private sectors in modern societies and it would be worth providing Ugandan academics, managers and bureaucrats with opportunities to see how these new structures are working and might be adapted to local situations. Just as statistics departments are collecting data according to what was new and exciting in the early 1970s, so too are bureaucrats trying to live in outdated organisational designs.

With regard to decentralisation the donor community has a responsibility for co-ordinating and supervising the provision of some funds at district level. The donor community also has a role in improving the effectiveness of resource use and influencing its allocation. One aid official said that district leaders seem to be genuinely committed to tackling the problem of poverty but that they do not know what to do about it. There is a role for donors and NGOs to help them to develop plans and policies which address development and social sector issues. He suggested that in the absence of Plans and Development Budgets it is likely that local politicians will encourage expenditure on large capital items such as football stadiums, and council halls.

6.3.3 Policies for equity:

In Uganda the World Bank has not yet shown a real commitment to poverty reduction as compared to its commitment to "growth". The donor community must maintain its pressure on the IFIs and the government if aid is to make any difference to poverty.

Decentralisation: Donors and NGOs should monitor the situation and ensure that aid interventions are equitably distributed with a focus on positive discrimination in favour of the very poorest districts. They should also be aware of the potential effects on political stability if the North and the East are allowed to lag further and further behind the West and Central Regions.

Donors have a responsibility for endorsing the message that corruption does not pay, and for making sure, that so far as donor funds are concerned, this is the case.

The establishment of law and order: With Western aid the NRM government is rebuilding a "cowed and withered" judiciary, while reforms to diversify and professionalise the army are progressing. Aid is also being provided for rehabilitation of the police service. If the North and East are not to fall further behind the rest of the country in terms of growth and poverty, it is vital that security be established as soon as possible. In particular threats from Kony's followers and from the Karimojong need to be dealt with; this requires both short and long -term action. Western governments should use diplomatic channels to try to control support and arms coming from outside Uganda.

The responsibilities of donors and NGOs: There is a lot of disillusion in the aid world at the moment, particularly in relation to Africa, and the time is ripe for all people working for multilateral and bilateral donors and NGOs to reflect on the goals, practices, structures and cultures of their institutions, particularly the unintended consequences of what they are doing. One problem is that the incentives provided by the aid industry for its employees tend to be related to such matters as money loaned or reports written rather than poverty reduced. Turnover in particular jobs is often rapid so that people do not have to live with the consequences of their decisions, and evaluation does not feed back into design and implementation of projects. Expertise in a particular country or region of a country is rarely allowed to develop.

Donor practices: Many development policies, programmes and projects are short-term, based mainly on restricted economic analyses, and reflect little awareness and understanding of the diverse and complex social and economic relations involved in the different farming systems and informal sectors of the Ugandan economy. They frequently adopt a top-down simplistic approach which comes to grief when it meets the realities of politics, organizations, emergent norms and cultural beliefs. While lip-service is increasingly paid to participatory and "bottom-up" approaches the organizational structures and cultures of Government and most donors means that, in practice, very few workable projects are designed, implemented and properly evaluated. If accountability is to be meaningful the Government, in conjunction with donors where this is relevant, should be required to publish evaluations of programmes and projects which have been publicly funded (including aid-funded). These should be used to inform the design of future programmes. At the moment the incentives for Government or donors to perform efficiently are very low.

We came across a number of examples of bad planning on the part of donors and will provide here just three examples. The first relates to the civil service reform. For the foreseeable future the government will not be in a position to fund proper pay for the civil service without external assistance. However, donors' projects tend to assume that the government, through improved pay, will be able to sustain the activities of the projects once donor funding stops. However, if people are not getting enough pay and are losing the "incentives" provided by the project it is unlikely that they will commit all their time and energy to continuing the project activities.

The second example comes from the field of education. The White Paper on Education which was agreed in March 1992 was a Ugandan initiative. The government followed up the White Paper with a list of proposed and prioritised projects but these were overtaken by the concurrent activities of donors driven by their own development timetables (Evans, 1994). Evans suggests that the long-term effects of such behaviour is problematic. "Donor-sponsored projects are like waves on the ocean, influencing appearances greatly in the short term, but not necessarily changing the larger swells on which they may ride". These reside in the beliefs and values of ordinary people and these are influenced by their long-term experiences. He suggests that "effective implementable policy may be that which comes closest to reflecting the values embedded in this underlying community of beliefs." (ibid, p21). Policy effectiveness should be evaluated over decades, not 5 years. "Donors, individually and collectively, indicate a desire to help Uganda strengthen its policy-making capability. Yet, the internal dynamics of the assistance process, driven as it is by timetables and forces within each donor, often work in ways which undermine or overshadow indigenous attempts to form and implement educational policies." (ibid, p22).

The third example comes from agriculture. When all the elements for a successful outcome are not included in a project or programme resources may be wasted. The first IDA funded project supporting sectoral adjustment, the Agricultural Rehabilitation project which ran from 1983 to 92, involved the physical rehabilitation of export processing facilities in cotton, coffee and tea, and this was successfully done. However, there was no provision for operational reforms with the consequence that not all facilities rehabilitated have been used profitably (World Bank, Agriculture, 1992, pxi)

The major problem with budget support is the difficulty of tracing where the money has actually gone. The major problems with project aid relate to the design of projects, and sometimes a lack of awareness or research related to the real needs and problems of the recipients. It is very hard to get resources to the poorest members of communities. The development of participatory approaches, (described below) should improve things although then donors will have to face the problem of whose agenda the aid programmes is really serving.

Donors have become increasingly aware that their organisational structures are not designed to implement participatory-style projects and are increasingly funding NGOs to undertake these. One problem arising from the increased donor interest in using NGOs to deliver services and projects is the corresponding rise of "briefcase" NGOs

NGOs practices: The World Bank report on the role of the NGO sector in poverty alleviation concluded that there was not much point in trying to define either an NGO or a CBG (Community-based Group) since they "are extremely diverse in nature, development ideology and self-perception" (NGO report, 1993, pi). They include policy advocates, community mobilizers and direct providers of services. The report suggests that while they have many strengths, they also have technical and managerial weaknesses. They usually work on a small scale and their projects may not be easily replicable or sustainable. "Often they are independent-minded, pursue their own agenda, and are not accountable to government, donors or to beneficiaries" (ibid, ii). There are a number of constraints which affect the efficient operation of NGOs including their organizational and technical capacities, government laws and regulations, and donor procedures. If these constraints were reduced they would be more effective (World Bank NGO Report, 1993, pii).

There is a problem related to competition among NGOs. The ActionAid Uganda document speaks of the threat of "the arrival of other major development NGOs in the same area with practices and policies significantly different from AAUs leading to confusion in community expectations". There is also a level of competition for donor funds, and for international NGOs the funds they raise in their home countries.

There is a need for intermediaries to link bottom-up structures to regional, national and international structures giving access to savings, technology and markets. Current links operate via the elites who dominate the politics and economics of rural areas. Macro and micro development must take place simultaneously and must be linked in rational ways.

Community-based groups (CBGs): Survival in rural Uganda has always depended on private enterprise except for schools, some health services and personal credit. People look to the family, clan, mission or PTA rather than the chief or the ministry official for moral authority and managerial capacity. The State is often seen as a problem maker: collecting taxes, licensing authority, enforcing controls.

Throughout Uganda communal responses to health needs can be found and could be built on; for example "muno mukabi" groups of friends who assist each other in time of need (UNICEF, 1994 p32). In the mountainous region of Kigezi local people have come up with an appropriate means of health transport within their means using local materials: the "engozi" or basket stretcher for hand carrying patients (ibid, p30). To put them to maximum use they have formed engozi societies for ownership and service (carrying). Almost everyone in the region belongs to an engozi society: most commonly non-members pay a hiring fee and members use it free of charge. Spontaneous community organizations can lack skills and resources and tend to focus on mobilizing people to provide labour. They could be helped to expand their role to identify health needs and prepare plans.

There are numerous other CBGs established locally and operating on a modest scale. They are usually ad hoc, established to meet a particular need, organized informally and with little experience or resources. Examples include credit and savings schemes among women and Hoe groups (women who hire their labour out during the harvest and weeding seasons for cash payment or payment in kind).

However, increasing community participation need not lead to projects which reduce poverty. Rural elites, who tend to try to control the participation process, may be more interested in projects of benefit to themselves. Poor and isolated farmers will need information and "sensitising" before they are able to participate in a meaningful way. Rural elites will also be male and it is important that women are brought into the decision-making and choosing process.

Sustainable development includes the requirement that communities have been empowered to initiate and manage community-based projects aimed at solving their problems and needs. A vital ingredient is local capacity-building involving such things as the training and equipping of community volunteers; sensitising children to importance of education; establishing a community accounting system; distance education. ActionAid Uganda pays an allowance to community and government volunteers. After decentralisation the regional tax base might provide supplementary funding to support the costs of local development intermediaries after the project has gone. Sustainability involves strict accountability systems into which community must be trained and the discouragement of dependency on outside resources.

Aid co-ordination: With respect to poverty and the distribution of aid resources the limited co-ordination between government, donors and NGOs makes it more likely that poor groups, districts, communities and people will fall through the net. Differing development styles between multilateral donors, different bilateral donors, NGOs, government at central and local levels, and people living in local communities can lead to problems of co-ordination and implementation. A crucial distinction is that between blueprints and processes. Too often discussion of policy stops with the former. It is important that the design of a policy be good enough in its own terms and implementable in the given environment. International NGOs need to co-ordinate and exchange strategies through regular networking and consultations with other NGOs and government departments. However, while lipservice is frequently paid to the need for co-operation and co-ordination the incentive structures facing multi-lateral and bilateral donors and international and national NGOs make it unlikely that this will be easily achieved. Even now there is no proper database in the country about which poverty issues different players are addressing. DENIVA (the Development Network of Indigenous Voluntary Associations) was established in February 1988 as an umbrella organization for local NGOs.It has attracted the interest and support of some donors but NGOs have felt it has no relevance to their needs. "While many of them expressed a need for more co-ordination among NGOs, none thought that DENIVA could adequately fulfil this role, despite it being one of DENIVA's original objectives" (World Bank, NGO Report, 1994). The Ministry of Internal Affairs, which is charged with registering, monitoring and advising NGOs has not co-ordinated activities because of resource constraints and neither has the External Aid Co-ordination Unit in the Ministry of Finance and Economic Planning with regard to international aid. Lack of co-operation among development agencies is exacerbated by lack of understanding, trust and respect on the part of government ministries, departments and officials, an inadequate policy framework, red tape, and outdated and inappropriate laws (Nakintu, 1994). One result is that the plentiful experience obtained by NGOs at micro level is not integrated into macroeconomic and social policies that affect the poor.

In addition to integration within projects and programmes there is a need for the closer integration of development policies, processes and agencies. "If external donor support is to be used in a capacity-building way for the benefit of the health sector, as opposed to simply operating a vertical project in isolation, government, donors and districts will need to formulate and implement effective strategies to avoid dependency and fragmentation. This will require planners, funders and implementers to utilize intersectoral thinking, as well as a perspective that includes the whole health sector. In this approach there is hope for rebalancing some of the demonstrated inequities in health." (UNICEF, 1994 p41)

One of the problems is culture differences between donors, the government, local government, NGOs and communities. "NGOs involved in PAPSCA and NURP have experienced continuous frustrations at the interface between the divergent institutional cultures which characterise NGOs, the Government of Uganda, and the Bank. Cumbersome procurement and administrative procedures slow NGOs down, as do important delays in the release of funds. (Goetz et al, 1994, p40).

So far we have described an unintegrated system of national and local government and an unintegrated system of donor and NGO aid. One of the major tasks of these systems is the reduction of poverty, but, once we try to put the 2 systems together it becomes hard to see how this is to be achieved in other than a patchy way and with great wastage. Greater integration will require political will on all sides and taking a longer-term view. When one does that it is clear that donors and NGOs should be doing all in their power to help the government at all levels to build structures, relationships, services and programmes which will last and which fit well with the social, cultural and political environments within which they have to work. This means allowing the government to produce its own plans, and prioritise what it wants, even if the plans and priorities are not always totally to donors' tastes. The alternative, where donors introduce projects quickly, outside the government's plan, as they recently did in the education field (Evans, 1994), is not likely to lead to ownership in the long run.

One of the jobs of the district authorities is to control and direct NGO presence. Decentralisation puts greater control at district level and NGOs will have to invest time and energy in collaborating more closely with district authorities; sharing information, plans, budgets and taking their views into consideration. A number of NGOs have been working with field-level government workers in an attempt to make them more efficient. For example, CARE has worked with extension agents in Bushenyi which has led to the enhancement of the quality of government extension services. They are about to begin the same kind of work with health workers in Mbale through an ODA-funded district health project (Goetz, p41).

Research into poverty is needed at all levels of Ugandan society. There is a danger that research by professionals is guided by the professional interests (both academic and promotional) of the researcher, rather than the needs of those being researched. The knowledge local people generate may be of benefit to no-one but the author of papers and books (Edwards, 1994). Knowledge also is a source of power and if it is concentrated in the hands of elites is less likely to be used in the service of poverty reduction. Another aspect of this kind of research is that local knowledge is devalued. It is therefore important to pay as much attention to the processes whereby knowledge is generated (and ways in which this might be done which are really beneficial to the participants) and disseminated and to involve all possible levels of Ugandan society.

For example, UNICEF believes it is important to get people to find out things for themselves. There are problems which do not become problems for action until they are pointed out or people become aware of their dimensions. So people may think the schools are fine but do not know what proportion of children in the community are attending. If local people are asked to do a survey to find out how many are at school it is brought to their attention and they may compare it with neighbouring communities and decide something should be done about it. UNICEF acts to guide discussions and information systems at local level. It is a process - to bring the facts out so that they know. The same could apply to expenditure on alcohol. There are RCs that ban the brewing of alcohol but this only means that it flourishes. Awareness of the extent of expenditure and the sacrifice of other expenditures can influence social behaviour via shame mechanisms and local pressure. Discussion of these issues at seminars in Kampala have a marginal effect while discussion by the people who are doing them is more likely to lead to change.

Poverty monitoring needs to be done in a purposive, integrated and "centralised" (as opposed to marginalised) way. The Ministry of Finance and EPD should take responsibility for setting up and co-ordinating a Poverty Monitoring Group which would involve the Statistics Department, the economics, sociology and politics departments at Makerere, interested Institutes and research centres (MISR, CBR, CEPR, the Women's Department), interested NGOs (eg ActionAid) and donors (eg the World Bank, UNICEF) and any other interested member of civil society. The Bank or another donor should be approached to finance local consultancies if necessary, research costs, dissemination etc. The group should develop an integrated programme of qualitative and quantitative research involving communities, households and individuals.

For monitoring purposes a simple very short household and individual questionnaire should be developed for administration to a sample of households (initially chosen randomly but then used as a panel) and individuals within the community (at monthly intervals to start with) at selected "sentinel sites" - rural and urban communities selected to represent the different farming systems, urban circumstances and cultures of Uganda. These questionnaires should be developed after a community research programme has been implemented. Information should be collected on various aspects of the community in a participatory manner. Topics covered should include economic activities - legal, semi-legal and illegal, links with urban areas, rules relating to access to land, sales, borrowing etc and recent changes, labour markets and other forms of labour exchange, social networks of support, social investment, seasonality, history, local definitions of poverty, a wealth ranking, infrastructure, schools, health services (public and private - including traditional), quality of RC administration, local rules applied that might affect economic activity, local corruption, quality of local judiciary, who are the "big men", patron-client relations, social investment, status-seeking expenditure, extent of alcoholic consumption, incidence of AIDS, malaria etc. This list should be supplemented with suggestions from members of the local community. Thereafter, each month certain community variables should be monitored (eg prices of outputs, inputs, rain, pests, attendance at school, diseases, availability of drugs, land sales, wages, quality of harvest, security problems, theft and murder, etc).

Attempts should be made to incorporate as many NGOs as possible into the research programme. They should be encouraged to work with local RC1s in the data collection process and results should be circulated widely in the area and district as soon as it is available as well as being sent to the centre for dissemination with the other results (also as quickly as possible). The emphasis should be on involving as many local people as possible in the data collection process and getting results out as quickly as possible. This kind of activity is something that NGOs can do more cheaply than anyone else, and if it is regarded as a longer-term capacity building and empowerment process for the communities it might become an accepted part of NGO work.

Once a network like this is established then policymakers and donors who are interested in the effects or possible effects of a certain policy or programmes will be able to use the research group to pursue the question.

The research group can use these "quick and dirty" findings to develop questionnaires and qualitative research in a more rigorous way, to explore the different facets of poverty (at individual, household, and district levels) and its causes. They should inform the activities of the Statistics Department and encourage multi-disciplinary co-operation in research, particularly within the University. Where possible students should be involved in the fieldwork to sensitise the future elite to problems of poverty.

If the political process produces some consensus that Uganda needs a Poverty Reduction Strategy then poverty has to be a major criterion in the development of all policies and programmes (although it need not necessarily always win) and an integrated programme needs to be developed. So, for example, the Government should initiate and contribute largely to the design of the next Poverty Assessment, while the major input should come from local consultants, with assistance from outside only when necessary. The Poverty Research Group described above should play a major role, together with appropriate people from national and district government and civil society. The aim should be to consult and discuss with as many local people as possible. The next World Bank Public Expenditure Review should have the real impact of public expenditure on poor groups as a major focus. All policies, loans and project aid should be assessed in terms of their projected impact on distribution and poverty. If it is deemed that there is none or it will be detrimental in increasing poverty the case for continuing must be strong and consideration given to possibilities of alleviating the plight of those being made worse off.

Institutionalising poverty awareness is always difficult, as can be shown by the fate of PAPSCA. It is vital that the task be given to a central ministry (MFEP) and to someone high in the hierarchy.

Government, donors and NGOs need to expand their research and their discussions to consider how a number of Uganda's cultures and related institutions actually work. Particularly relevant are the "grassroots" cultures (which vary between ethnic and other groups), the business culture, the political culture, the civil service culture, and other organisational cultures. The problem of coping strategies which have detrimental effects on growth or longer-term poverty needs to be addressed, while the need to take steps towards the integration of formal institutional structures with surrounding societal structures cannot be stressed enough.

ActionAid Uganda, Country Strategy Paper, 1994

Alexander, J et al, The Micro-macro Link, University of California Press, Berkeley, 1987.

Ashfar, H and C Dennis, Women and Adjustment Policies in the Third World, Basingstoke, Macmillan, 1992.

Appleton, S, "Problems of Measuring Changes in Poverty over Time: the Case of Uganda 1989-1992", Oxford, CSAE mimeo, 1994a.

Appleton, S, "'The Rich are just like us only Richer': the Determinants of Poverty In Uganda", CSAE mimeo, 1994b.

Appleton, S, A Balihuta, P Bevan,, J Mackinnon, A Ssewaya, G Ssemogerere, Poverty in Uganda in 1994: a research report, CSAE mimeo, 1995.

Balihuta, A M, and G Ssemogerere, "The Determinants of Access to Education and their Implications for Poverty Alleviation in Uganda: analysis of the IHS Data 1992/3 and related studies", Department of Economics, Makerere University, 1994.

Bangura, "Economic Restructuring, Coping Strategies and Social Change: Implications for Institutional development in Africa", Development and Change Vol 25, 1994.

Barnet, T, and P Blaikie, AIDS in Africa; its Present and Future Impact, London, Belhaven, 1992. 

Bates, R H, Beyond the Miracle of the Market, Cambridge, CUP, 1989.

Bassett, T and D Crummey, Land in African Agrarian Systems, London, University of Wisconsin Press, 1993.

Berry, J, Y Poortinga, M Segall, and P Dasen, Cross-cultural Psychology, Cambridge, CUP, 1992.

Berry, S, Cocoa, Custom and Socio-economic Change in Rural Western Nigeria , Oxford, Clarendon Press, 1975

Berry, S, No Condition is Permanent: the Social Dynamics of Agrarian Change in sub-Saharan Africa, University of Wisconsin Press, Madison, London, 1993.

Booth, D "Development Research: from Impasse to New Agenda" in F J Schuurman (ed), Beyond the Impasse, London, Zed Books, 1993.

Booth, D, Social, Economic and Cultural Change in Contemporary Tanzania: a People-Oriented Focus, Report to SIDA, 1993.

Booth, D, " Rethinking Social Development: an Overview" in D Booth (ed) 1994.

Booth, D, (ed) Rethinking Social Development: Theory, Research and Practice, Harlow, Longman, 1994.

Booth, D, "Institutional Reform and Adjustment: Perspectives from Social Theory", Paper presented at the Fourth Structural Adjustment Forum Workshop, Nottingham, January 1995.

Boudon, R, Theories of Social Change, Cambridge, Polity Press, 1986

Bourdieu, P, The Logic of Practice, Cambridge, Polity Press, 1990.

Bratton, M and B Liatto-Kiandu, "A Focus Group Assessment of Political Attitudes in Zambia", African Affairs, 1994

Brett, E, Providing for the Rural Poor: Institutional Decay and Transformation in Uganda, IDS, 1992; Kampala, Fountain Publishers, 1993.

Chambers, R, Rural Development, Harlow, Longman, 1983.

Coleman, J and T Ferrara , Rational Choice Theory, London, Sage, 1992.

Edwards, M, "Rethinking Social Development: the Search for 'Relevance'" in D Booth (ed) 1994. Ensminger, J, Making a Market, Cambridge, CUP, 1992.

Evans, A, "Growth and Poverty Reduction in Uganda", paper given at the Conference on Poverty Reduction and Development Co-operation, Copenhagen February 1994.

Evans D R, "The Evolution of Education Policy in Uganda: Continuity and Change" in (ed) Hansen and Twaddle, forthcoming.

Feierman, S and J Janzen, The Social Basis of Health and Healing, Oxford, University of California Press, 1992.

Gladwin, C (ed), Structural Adjustment and African Women Farmers, University of Florida Press, 1991.

Goetz A M, S Maxwell and H Manyire, SPA Poverty and Social Policy Working Group Consultancy on Poverty Assessment and Public Expenditure Issues: Uganda Country Study, Kampala and Sussex (IDS), 1994.

Goldthorpe, J H and P Bevan, "The Study of Social Stratification in Great Britain : 1946 - 1976", Social Science Information sur les Science Sociales, 1977.

Hansen, H B, and M Twaddle, Uganda Now: Between Decay and Development, London, Currey, 1988.

Hansen, H B, and M Twaddle, Changing Uganda: the Dilemmas of Structural Adjustment and Revolutionary Change, London, Currey, 1991.

Hansen, H B, and M Twaddle, Developing Uganda, forthcoming

Hill, P, The Migrant Coco-Farmers of Ghana, Cambridge, CUP, 1963.

Hill, P, Dry Grain Farming Families, Cambridge, CUP, 1982.

IIES, RRA Notes 1-19

Holton, R, Economy and Society, London, Routledge, 1992.

Janvry A de, E Sadoulet and E Thorbecke, "Introduction" to ILO Conference Papers on state, market and civil organizations in rural development in World Development, 21.4, 1993

Knight, J, Institutions and Social Conflict, Cambridge University Press, Cambridge, 1992.

Knorr-Cetina, K, and A V Cicourel, Advances in Social Theory and Methodology; toward an integration of Micro- and Macro- Sociologies, Routledge and Kegan Paul, London, 1981.

Kontopoulos, K, The Logics of Social Structure, Cambridge University Press, Cambridge, 1993.

Langseth, "Civil Service Reform: Objectives and Strategic Plan" in (ed) Hansen and Twaddle, forthcoming.

Layder, D, New Strategies in Social Research, Cambridge, Polity Press, 1993.

Long, N, Social Change and the Individual: Social and Religious Responses to Innovation in a Zambian Rural Community, Manchester, MUP, 1968.

Long, N and A Long, Battlefields of Knowledge, London, Routledge, 1992.

Long, N and M Villareal, "Exploring Development Interfaces: from the Transfer of Knowledge to the Transformation of Meaning", in F J Schuurman (ed), Beyond the Impasse, London, Zed Books, 1993.

MacGaffey, J, Entrepreneurs and Parasites: the Struggle for Indigenous Capitalism in Zaire, Cambridge, CUP, 1987.

Mackinnon, J, "The determinants of Child Nutrition and Mortality during the Political and Economic Recovery in Uganda", CSAE mimeo, September 1994.

Mouzelis, N, Back to Sociological Theory; the Construction of Social Order, Basingstoke, Macmillan, 1991.

Munch, R " The Dynamics of Societal Communication" in ed P Colomy, The Dynamics of Social Systems, London, Sage, 1992.

North, D, Institutions, Institutional Change and Economic Performance, Cambridge, CUP, 1990.

North, D, "The New Institutional Economics and Development", Paper prepared for the Conference on Public Choice and Development: the New Institutional Economics and Third World Development, London, September 1993.

Oppong, C, Female and Male in West Africa, London, Allen and Unwin, 1983.

Scott, J, Social Network Analysis London, Sage, 1991.

Sender, J, and S Smith, The Development of Capitalism in Africa, London, Methuen, 1986.

Serpell, R, The Significance of Schooling: Life-journeys in an African Society, Cambridge, CUP, 1993.

Seur, H, "The Engagement of Researcher and Local Actors in the Construction of Case Studies and Research Themes" in Long and Long (ed), 1992.

Sender, J and S Smith, Poverty, Class and Gender in Rural Africa: a Tanzanian Case Study, London, Routledge, 1990.

Ssemogerere, G, "The Impact of Macroeconomic Policy on Poverty in Uganda 1989 - 1992", Department of Economics, Makerere University, 1994.

Stigler, J, et al, Cultural Psychology, Cambridge, CUP 1990.

Thompson, M, R Ellis and A Wildavsky, Cultural Theory,. Oxford, Westview, 1990

Thorbecke, "Impact of State and Civil Institutions on the Operation of Rural Market and Nonmarket Configurations", World Development, 21.4, 1993.

Toye, J, Dilemmas of Development (2nd Revised edn), Oxford, Blackwell, 1993.

UNICEF, Uganda National Situation Analysis, Kampala, 1994

Uphoff, N, "Grassroots Organizations and NGOs in Rural Development: Opportunities with Diminishing States and Expanding Markets", World Development, 21.4, 1993.

Vaughan, M, The Story of an African Famine, Cambridge, CUP, 1987.

World Bank, Structural Adjustment and Poverty: a Conceptual, Empirical and Policy Framework, 1990

World Bank, Growing out of Poverty, 1993a

World Bank, Agriculture, 1993b

World Bank, Social Sectors, 1993c

World Bank, NGO Report, 1994

This Appendix makes use of secondary sources to explore how the different dimensions of poverty affect identified poor and vulnerable groups in Uganda and the coping strategies they adopt to survive and escape from poverty. It relies heavily on two reports that have recently appeared: the UNICEF Situation Analysis for 1994, and the ActionAid Country Strategy Paper 1994-199 and its background papers. The groups are considered under four headings: class poverty, status poverty, contingent poverty and poverty related to particular policies or policy-related activities such as war. (See Fig 1.1) It is important to be aware that an individual may fall into one or more of the categories and that categories may contain people who are not poor according to narrower economic definitions (eg PWAs) . One of the reasons for identifying these categories is that they are particularly relevant for identifying the kinds of policy intervention that might be made (eg legislation, public services, fiscal policy, focused projects, emergency aid etc). The groups might also be used to assess the affects of different policies on the poor and to monitor them over time. Some of these groups will be concentrated in particular areas or regions.

Resource-poor farmers are part of the larger class of the peasantry composed of rich, middle and poor peasants. They have limited access to productive resources, their basic social services are inadequate, and they cannot produce enough to feed their families properly throughout the year. Housing conditions are extremely poor. The resources they lack include capital, labour, enough land of reasonable quality, technology, education and skills, health services, transport, and access to credit. They live in villages with rich and middle peasants and are often dependent on them for access to land and cash leading to the possibility of exploitation.

Figure .2 is taken from a wealth-ranking done by ActionAid in Mitiyana in Mubende District. ActionAid believes that Kalagi is the more typical village. Mamdani has described life for the very poorest of these groups. "Those individuals who live in mud huts, the majority with grass thatched roofs, a minority with iron sheets crying out for urgent repair or replacement. They sleep on papyrus or on grass at times or on bare ground. Seldom do they have a second change of clothing or night covering. Quite often in the rainy season when temperatures are lower than average, the same piece of cloth that the wife wraps round herself during the day is what the couple uses for a sheet to cover themselves at night. It is people in such situations who most often resort to drinking cheap but strong brew that also serves as a sleeping sedative" (quoted in ActionAid Uganda, 1994).

The following descriptions of the wealth categories were provided by the villagers themselves. They also did the ranking.

Accessibility to agricultural extension by the resource poor is very minimal; in many areas government extension staff do not have the means let alone the morale to reach the peasants in the villages. The few who manage tend to work with either the middle or rich peasants because they have the means of production and can get results and be motivated. In Kabale there are not enough field staff (1 extension staff for 15,000 households) and a lack of transport, and the situation in other areas of Uganda is similar. This problem also affects NGO activity. "The issue of working with the poorest sections of the community is still unresolved. Many agencies have tended to work with the middle and rich peasants rather than the poor ones." Poor peasants are shy, do not turn up for meetings, and have limited access to factors of production. The ActionAid project in Mityana has been operational in the area for more than 10 years and "is still thinking of ways and means of availing its agricultural extension to the poor peasants. It is now identifying the poorest groups and has plans to develop specific interventions for them.

The landless are defined as having no access to land or insecure tenure. They are often rural labourers on large farms or plantations or subsist in small rural towns and market centres through petty trading, crafts, paid or casual employment. "It is difficult to quantify the problem, but recent estimates put the figure of the landless at 2.5 million. Population pressure and the customary inheritance system has resulted in land fragmentation and landlessness in areas such as Kabale; but the problem is widespread and increasing" (ActionAid, 1994 p17). Many people are potentially landless and therefore vulnerable as a result of government land laws, population pressure and internal strife leading to displacement.

The land-poor: Today there is borrowing (or share-cropping) of land by those who cannot afford to buy it. The current legal and institutional framework does not provide for the equitable distribution of land or security of tenure for most cultivators. Tenants on public land (most cultivators) are defined by law as tenants on sufferance. Leasing of public land can only be afforded by the rich and in some places the 1975 Land Reform Decree is being used to gain leases; land prices are rising. Customary land tenure leads to fragmentation. Some rural people may have access to land but not enough to meet subsistence needs.

Squatters on gazetted land (forest and game reserves) have been evicted and in some cases resettled while others are in transition. Squatters on mailo land are tenants on sufferance and could be evicted and sometimes are. These are people who fall within the land that was given to the Kabaka and his chiefs according to the 1900 agreement. "Many of these have already been evicted after bitter struggle and sometimes fatal confrontations."(ActionAid Uganda L, p6) They often live in very poor housing as they have no incentive to build good houses. The government does not provide services eg health etc. The practice of borrowing or renting land leads to exploitation by the rich while small amounts of land lead to over-cultivation and soil degradation.

Projecting into the future the UN Human Settlements Study says that given the population size and the projected growth rates there is not enough land available in large enough units to provide most farmers with the opportunity to increase production if productivity is not raised. The current amount of land available for cultivation per person is 1.07 hectares. If the population reaches 23 million by the year 2000 the available amount of land will be 0.79 hectares per person (ActionAid, 1994 (UNICEF, 1994, P4)).

New settlers have usually moved out of areas of extreme land pressure and fragmentation and are very vulnerable. They face a number of problems. They have to search for new places of settlement, acquire new property, and gain acceptability among the people already in the place where they settle. There are two categories of new settler: the voluntary and involuntary. The latter usually move in a group usually to a designated place. Voluntary movement has long history in Uganda. In Bacilli County Mubende district it is estimated that 80% of population are recent settlers or children of recent settlers. The Women Needs Assessment Survey of 1992 found that 58% of women in district were born outside it. The reasons include a search for land, avoidance of social conflicts, a search for employment opportunities, family ties, accessibility to social services, and getting away from gazetted areas. The majority move as a result of land pressure. (ActionAid, 1994)

The existence of large areas of unused land is no guarantee that people moving will be able to secure it, especially where it is communally owned, eg in the North and Northeast. In Buganda it is easy to buy and acquire land though there may be problems with insecurity of tenure. People tend to move to areas with similar climates and cultures. Some places are now habitable after tsetse control; in parts of Kibale, Hoima, Masindi, Kamuli, and Tororo.

Historically new settlers migrating from Kisoro, Kabale, Mbarara, Bushenyi and Bugisu have mostly become worse off at their destinations. They have tended to settle in areas where labour is in great demand; especially in Buganda or on the plantations. They have also moved to districts with vacant land (eg Kibale, Kabarole) even though the new areas may not be particularly productive. During the first years of settlement new settlers face food shortage, poor health, inadequate housing, and limited access to productive land. Sometimes they are not readily accepted in community and for this reason one tends to find one ethnic group concentrated in a given area. Over time they acquire enough land to improve the situation but often not enough to enable them to meet basic family requirements. They cope by getting food from relatives or selling labour.

The poor ones, coming with little or no money, cannot buy land and often settle on marginal land which maybe gazetted. Production is low and they are unable to improve soil. They work for rich landlords, becoming vulnerable to eviction and exploitation. They are not denied access to the social and economic facilities in the community.

Because of large numbers communities affected by the arrival of non-voluntary groups of settlers are thrown into a panic as no-one seems to prepared to take responsibility for them creating an atmosphere of insecurity. There is often sympathy but local communities do not have much to offer. ActionAid quote the example of Kibuye village in Luwero District where about 3000 people are just settling after being evicted from Kiguuli village because it had been set aside for the development of government projects. The arrivals usually had 1 meal a day provided by a relief agency. Sanitary conditions and infrastructure were poor. the people had low income or no income, and little land was available for them.

The incomers have a longer-term impact on the residents often involving intermarriage, the use of a new language mixture, changes in cultural attitudes, practices and beliefs. The increase in population leads to increased pressure on inadequate resources, and often soil degradation and indiscriminate bushclearing. However because more land is cleared there may be a decrease in vermin, and new trading centres, markets, and crops may develop.

Pastoralists: The total number of pastoralist holdings in Uganda (excluding Karimoja, Kumi and Soroti) was estimated in 1990 to be 150,000 with over 3 million head of cattle. ActionAid estimates that Karimoja, Kumi and Soroti contain at least the same number of households and more cattle. Pastoralists used to be concentrated in the southwest and north east, but there has been movement to central areas, particularly Mubende, Kiboga and Luwero (ActionAid, 1994 p17).

A poor pastoralist has been described as one who owns less than 50 head of cattle, squats, grazes communally, cultivates less than 1 acre of land, does not own a water source, and has no farm structures (ActionAid, 1994 p17). Cattle rearing is part of the culture; income opportunities are limited, marketing facilities for livestock and livestock products are poor and climatic and cultural conditions impair consistent production. Imposition of quarantine against contagious diseases disrupts market outlets. Low animal productivity is the result of diseases and pests. Veterinary extension services are available but inadequate, particularly where mobile pastoralists are hard to find. Pastoralists have particular problems deriving from poor communications and resources, mobility, lack of security of tenure, and lack of access to services. Even if they are wealthy they are always vulnerable to losing everything as a result of drought or raiding.

Hunter-gatherers: There are a small number of hunter-gatherers in the south-west of Uganda. In Bundibugyo, ADRA (an NGO) has settled some at Burondo, in Kasitu sub-county. They gave them goats (New Vision claims this is the first time for a Bundibugyo pygmy to own an animal) hoes, pangas, seeds, and clothes.

The labour-scarce are those with high dependency ratios and few adults or children of working age. The groups includes households headed by old people, women, and PWAs who are considered below.

The urban poor are less numerous but the rate of urbanization (estimates range from 4.8% to above 9% with the rate in Kampala being higher) is growing steadily and their poverty problems are likely to grow in scale and severity.They live in squalor with poor housing conditions and a lack of sanitary and water facilities. Education levels are low, health and nutritional standards are poor, and income opportunities limited. Urban planning and provision of services has not kept pace with growth. Slum housing is particularly bad as most structures are unplanned and temporary; they are usually made of mud and wattle and roofed with papyrus mats or scrap tin sheets. They are built very close together with no space for compounds or latrines; most lack safe water, have poor drainage systems, are surrounded by stagnant ponds, flood during the rainy season, and have no electricity. The urban poor are often located in swampy areas, eg around Kampala. There are a large number of slums in Kampala including Katanga, Nakulabye, Kivulu, Bwaise, Blueroom, and Ndeeba.

The construction industry is booming but thousands still lack accommodation and "apparently sleep either in front of shops, at the railway station, in car garages or in goods containers (which are often used as shops during the day). The increase of city rents and lack of sufficient accommodation has led to the expansion of slums with houses being built in swamps." The most popular slum is Katanga where construction is increasing with or without permission. Most of these houses are not fit for habitation and are resulting in ecological and health problems.

The majority of the very poor are tenants often paying high rents. Government planning leads to insecurity. Land conflicts between government, local authorities, mailo owners and would be developers increases this insecurity. A large proportion of urban population live in 1-roomed dwellings with households sized between 4 and 5 (68% of houses in Kampala are 1-roomed) and most urban houses are shared and rented. Only 2% have water and electricity; 92% have neither. ActionAid pinpoints a number of constraints on the provision of urban services including lack of consultations with the poor, uncoordinated policies and implementation, and the unclear status of most poor communities which are often illegal.

The majority of the urban poor are unemployed, casual workers, or in petty trade. Many urban poor work in the informal sector where employment is irregular and unstable and earnings are often below subsistence. ActionAid gives the causes of urban poverty as unemployment, temporary employment, unpredictable employment, unstable employment, low educational levels, and exploitation by employers including underpayment and unregulated conditions of work. There are no policies governing casual and unskilled labour.

Slumdwellers often apathetically accept the situation, do what they can with what they have and experience little or no control over their surroundings (Bagonza, 1987). Water and sanitation co-ordination are poor, infectious diseases are rampant and the Infant Mortality rate is high. Unemployment is a widespread problem so slum dwellers resort to illicit businesses, for instance prostitution, selling stolen goods, and beer selling. The health services are inadequate. Paradoxically, most of the unemployed slum dwellers seem to think that if social services are improved, their expenditures will increase, especially on taxes or through repairs and maintenance of the amenities. They also think that if the place remains in its current condition it will continue to provide a ready market for stolen things. Also, if the houses are poor they will not attract more important people to the area who are a threat to the slumdwellers, while if the houses are improved they will have nowhere to go during the construction period and the new improved houses would be too expensive for them.

Women as a category suffer from a range of deprivations the most basic of which are social and cultural. Women suffer from asymmetry in the legal system and customary practices (under customary law and practice women are minors without adult legal status or rights) and de facto discrimination. These social and cultural deprivations are related to political, economic and personal deprivations; all five dimensions of deprivation interact with and compound each other. There are also particularly vulnerable groups of women: the disabled (estimated to be about 10% of all women); widows (no estimates); illiterates (66% of women); female-headed households (approx 1.8million) and HIV+ women (estimated to be 10-20% of women). Women fulfil multiple roles as home managers and clothes-washers (88%of activity), cooks (86%); child bearers and rearers (62%); carers (62)%; food producers; water collecters (70%); and firewood collecters (73%) (Growing Out of Poverty, 1992)

More than half the population (8.6m) are women and they are responsible for more than 75% of agricultural production. They are disproportionately found in the other poor and vulnerable categories. Two thirds of women are illiterate. They work an estimated average of 15 hours a day and are subject to social and physical violence and nutritional vulnerability due to food taboos and food insecurity (ActionAid Uganda, p16). The most vulnerable are female heads of households, widows and single mothers. Female headed households are often stigmatized because men resent their economic independence and the prevailing culture requires women to be married and subordinate. Women have been given little chance to control the major factors of production except for their labour and even that is not fully in their control. Bridewealth payment is still upheld by most cultures subjecting girls to early marriage and lack of access to higher education. It is defended by some cultures as a useful institution that promotes family cohesion, but it has tended to give women the status of a commodity.

Personal and physical deprivations: Women suffer a range of what might be described as personal deprivations. Multiple child-bearing and heavy labour leads to backache and other disabilities. Lack of investment in their human capital means they are more likely to be illiterate and uneducated than men. (Education for women is well-known to have beneficial effects in postponing marriage and the first child, and on children's welfare). They are also brought up to believe they are inferior to men which leads to lack of confidence. This makes it more likely that they will do badly at school, and makes them afraid to approach credit institutions for assistance because they lack technical knowledge and fear the repayment process (ActionAid Uganda, 1994, p8).

Economic deprivation: Under most customary succession and inheritance laws land properties are mainly acquired by men through inheritance. Without land women do not have collateral for credit restricting new income activities.. Of the 97% of women who have access to land only 7% actually own it (ibid, p7). A survey conducted in Masaka district found that about 90% of the men own land, while 92% of the women own no land (Lubega, 1993). Landed properties are mainly acquired through inheritance from parents (usually fathers) and women do not usually fall into this bracket. If a daughter remarries the land may get into the hands of someone foreign to the clan so Ganda and Nyoro inheritance of land is strictly for boys. Sons get a greater share of the more valuable properties like land, coffee plantations, cows, roofed houses. Children of "illegal" wives often don't inherit anything. In Kigezi husbands distribute land to their wives and children which causes smaller holdings and land fragmentation.

Although the national law establishes that widows should inherit 15% of the property of a dead husband in practice customary law usually prevails and widows often inherit nothing; rather they are inherited by the husband's male relatives. Men will fight to inherit a widow whose husband has left property while women may be abandoned with her children if the man dies poor.

The written law is not much better. If the husband dies intestate the widow is only entitled to 15% of deceased's estate: 75% goes to the heir and children, 9% to relatives and 1% to the customary heir (Uganda Succession Amendment Decree 1972). Section 2 of the Succession Amendment Decree is clearly discriminatory. When someone dies without a will the following conditions apply:

Men predominantly make the decisions about food crops to be grown, including the amount of land allocated; while women and girls are the ones most directly involved in all aspects of food production, men may decide to allocate more land to cash crops at expense of food crops. Figure 3 shows the division of labour between men and women in a village in Arua as reported by the women.

The women said that at times the men invite communal labour to open up an ambitious venture then leave the remaining work for the women. One of the greatest constraints on the expansion of food production is the lack of additional labour; women's workload is already too heavy. For example, maize grain is hard and needs machines to grind it but there are not many available. Planners must be reminded that women are over-employed, and are not an economically idle resource awaiting supply response incentives. Time constraints have implications for the participation of women in economically productive and social sector investments and programmes. Thus mothers may have no time to take children for immunizations, no time to boil water and prepare special food, no time to go to literacy classes.

Women do 50% of the local trade and their income-generating skills include basket weaving, beer brewing, embroidery/needlework, and the sale of cooked foods. Semi-structured interviews with key informants in Mubende and Kibaale revealed that women have control over the sales of products like handicrafts and food which fetch little money. Men regard women and children in part as wealth objects so whatever labour they put into land is controlled by men and so are the proceeds. Permission to sell the women's products has to be sought from husbands. In most cases husbands do the marketing and sometimes withhold returns or report a false price. Women end up selling cheaply to middle men who come to the village. ActionAid found that in Mubende and Kibaale women are increasingly responsible for meeting household needs once paid for by men's cash crop income; 30% of women with children in school paid all or part of their expenses. Accordingly women are adopting strategies to generate income which include the sale of food crops, alcohol, labour, sex and other services.

Patterns of household expenditure reflect the status, bargaining power, and options of the members. In the north within the household women are visibly more poor than men; children describe their mothers as poorer than their fathers. In maleheaded households the share of alcohol in total spending is about twice that observed in female-headed households (Mission estimate using the Integrated Household Survey data).

Educational enrollment is substantially lower for girls, as is shown in Chapter 4. In villages in Mubende and Kibaale ActionAid found that education is considered to interfere with the traditional values of a good wife leading to loss in bride wealth. Parents tend to think that girls are by nature intellectually inferior to boys and therefore it is not worthwhile to educate them. It is prestigious to have highly educated sons. Illiteracy makes women vulnerable to scorn; parents tend to pull girls out of school to help with house and farmwork; school dropout rate is high in P6 and P7.

In rural areas there is an estimated maternal mortality rate of 600-1000 per 100,000 births which is one of the worst rates in Africa. In Madi county it was mentioned that women are reluctant to take babies for immunisation because it is believed that it causes serious diseases and death. It is believed that immunisation for older ages sterilises young women (ibid, p3).

In Vuu village the over-drinking of beer causes ill health in women. Women drink too much after work to achieve a psychological relief a trend that is increasing. In the past men used to stop this but now the two are the same. Women drink from the marketplace and at home in the late evenings; this changes people's morals and can lead to family breakdown.

Women's lack time, education, and information has effects on their children's wellbeing. Dirt and diarrhoea lead to lack of nutrition which is compounded when there is lack of time to cook for the baby.

Cultural deprivation: A very high value is attached to childbearing. In 1968 the lifetime fertility average of women was 7.1 births per woman and in 1991 it is still 7.1. 67% of women have had at least 1 pregnancy by the age of 18. A national family planning policy is only recently coming into being. UNICEF suggests that the reasons for the high fertility include a young age of marriage, low numbers of single people, minimal use of modern contraceptives, the low status of women, poorer educational opportunities for girls, the importance attached to having children as a source of labour and prestige and family members to look after one in old age. Most clinics distributing contraceptives are in urban areas while 80% of women of childbearing age are in rural areas. UNICEF suggests that up to 30% of births are mistimed and 5% unwanted (UNICEF, 1994, p45); the total number of births per lifetime is higher than the perceived ideal. In a 1988 study 84% of women had a favourable attitude to contraception while up to 40% or more husbands disapproved of it. The decision as to how many children is the husband's "When you want surgical contraception, the husband has to decide. Why should a man decide for me? It is my life." (Kiboga focus group) (UNICEF, 1994 p45). In Mubende and Kibaale medical family planning methods are believed to have serious side effects on users in the long run (ibid, p13). Courses in family planning have been directed at women so men fail to see problem.

Girls are less valued than boys and are given away at an early age in exchange for wealth. They are restricted from the cash sector and often hand over cash to their husbands. They are more often paid in kind, usually salt, soap or food. Cash earning brings higher status and men are not keen on women who are economically independent. In Kampala women traders are called Dubai which means "a woman who will do anything for the sake of earning money." Husbands sometimes make it their right to decide whether or not a wife should work for pay and in some cases men abdicate responsibility for families when their wives earns cash.

"Women have no choice about marriage; it is compulsory for a woman to marry. For a woman it is a curse not to marry" (Kabale focus group - UNICEF p55). Less than 2% of women never marry at any point in their lives. Civil marriages, religious marriages and customary marriages are all recognised by law. Christian church marriages are monogamous, Muslim marriages and customary marriages may be polygamous. There are also less formal consensual unions without legal ratification, religious ceremony or brideprice; if she has his child she becomes his "wife". There is a legal minimum age of 18 for marriage but customary marriages may be arranged for much younger adolescents, especially girls. Thirteen is considered acceptable in many areas. "Girls in families are forced to marry at an early age for getting cows so their brothers can get married"(UNICEF, 1994,p55). 41% of girls aged between 15 and 19 are married. "A recent study on reproductive health in Mbale, Mbarara, and Gulu found that over the last two decades the age of first marriage had been decreasing from 18 down to 14. They attributed this to hardship, war and insurgency, and a break-down of socio-cultural institutions." (UNICEF, 1994 p55)

Customary marriage is a binding of two families; the amount of brideprice varies by ethnicity and region. The following tribes still practice it: the Kiga, Nyankole, Bamba, Itesot, Nyoro, Lugbar, and Alur. It often leads to early marriage for girls as parents wish to obtain cows. Among many eastern, northern and western tribes brideprice tends to be high and paid in cows - if marriage ends in divorce it has to be repaid. In Central region brideprice is very modest and involves gifts, usually locally-brewed beer, with no question of repayment. In Central region brideprice is less of an obstacle to divorce or polygamy. "In Mawagola a father may demand 100,000 for his daughter. The youth works hard and may pay it off in instalments" (Masaka focus group) (UNICEF, 1994 p56). Not paying brideprice leads to a loss of prestige."In cases where brideprice is not paid, women are mistreated and considered to be cheap prostitutes" (Apac) (UNICEF, 1994 p56). Brideprice does not buy a woman but compensates her family for the loss of her labour. It tends to make marriages more stable as families negotiate to settle disputes. However in the UNICEF study district personnel said it was commonly agreed that brideprice meant that the woman belonged to the husband and his clan, in some cases as chattels (property). Brideprice entitled the man to the woman's labour, obedience, sexual availability, and fertility (UNICEF, 1994 p56).

In Bundibugyo the Bamba men (fathers and brothers) negotiate early marriages for girls limiting their opportunities in other aspects of life, such as education. The cows are used for the education and bridewealth of their brothers. In Vurra village in West Nile the elders said that a woman is supposed to produce as many children as the cows paid in bridewealth, and is expected to continue producing as the cows paid in bridewealth multiply. Bridewealth keeps women in marriages despite maltreatment in case parents can't pay back. Because of this a girl's natal family sees her as a "visitor" who will grow up to belong to another clan (UNICEF, 1994, p51) In marriage a woman is often regarded as a possession of the man or property of the clan but not a clan member.

Circumcision of girls is practised in Kapchorwa by the Sebei. It is a source of problems including bleeding, infection, and AIDS risk, and later reduced sexual pleasure, difficult deliveries and maternal mortality. (UNICEF, 1994 p52, ActionAid, 1994, p8). ActionAid report an interview with a Sebei clan head who explained that circumcision is "part of a rite of passage, a turning point from childhood to adulthood. Without the rite a woman would be regarded as childish, irresponsible, outcast, and importantly impure."

In most areas men who can afford to practice polygamy. "Every man has at least 2 wives" (Iganga interview (UNICEF, 1994 p56). It brings status and prestige, provides free labour, or may be a response to barrenness in the first wife or the lack of a male child. Multiple wives guarantees sexual access to a wife at all times (there are cultural taboos during menstruation, after a birth, and during mourning). In rural areas polygamy is more common among the wealthy and the educated. Some polygamous marriages are formal, especially among the Ankole. Wives live together under one roof or in the same compound. In less formal polygamous unions a man may have one wife in one village and one in another or in the city. Many men in the Central region maintain one or more extramarital relations and some groups look on them leniently especially if there are children. 33% of married women in rural areas and 31% in urban are in polygamous marriages; women with high education are less likely to join polygamous unions.

The Bahima practice wife-sharing; the husband has first priority in relation to a wife but she is considered an addition to the clan and the family and she is available sexually to blood brothers, relatives and friends. "District personnel report that wife sharing still occurs for this Western tribe, but it seems to be decreasing." (UNICEF, 1994 p57).

Within marriage sex is expected and obligatory for women; many people say it is predominantly for reproduction but there is also a commonly held belief it is necessary for good health (UNICEF, 1994 p58). "Women in the districts were articulate about their lack of voice within the family about sex" (UNICEF, 1994 p58). Extra-marital sex is the expected norm for men and up to 80% of rural women say their husbands have outside partners (UNICEF, 1994 p58).

Marital instability results from infertility, alcoholism, extramarital sex, a young age at marriage, polygamy, perceptions of laziness, physical abuse, economic problems, neglect, extended family obligations, and sexual dissatisfactions (UNICEF, 1994 p58). Divorce leads to economic deprivation for women and children; they can be deprived of material property, lose access to land for cultivation, have to repay bridewealth, and might have to leave children with the husband. Men only have to establish one ground for divorce, while women have to have more than one ground. She usually returns to live with relatives. Single women with children are stigmatized and often viewed as prostitutes.

The inheritance of the widow and children by a brother of the deceased is practised by some tribes in Uganda including the Banyankole. It is used to protect woman and children within the clan. The practice can increase the transmission of HIV/AIDS. Traditional practices are now changing; the widow has an increasing choice of partner among potential heirs and even whether to marry again. Some areas are choosing to support the widow and children without direct inheritance. Sometimes widows are just sent away. "Mbattude 42 once a resident of Lwezo village, Butambala County, Mpigi District narrates her ordeal with her father-in-law. "I lost my husband five years ago. He left me with 5 children, 1 boy and 4 girls. Recently my father-in-law held a meeting with other family relatives and they said that they can no longer support me. They have decided that I go back to my parents' home with the 4 girls and they retain the boy as he is more useful to the family. Our house has been pulled down leaving us with no shelter. I have been married for 11 years and have grown food on which I am feeding my children. I am not sure how I will bring them up" (UNICEF, 1994 p7).

Women are subjected to religious condemnation for causing the fall of man and hence original sin. Many religious practices marginalise women; there are no women priests and in some cases they must worship separately from men. Men tend to dominate religious committees and offices such as church warden. Most muslim families in rural areas do not send their daughters to school.

Women have less access to information than men. While the number of newspapers has increased over the years only 44% of women can read. State-owned Radio Uganda broadcasts for 18 hours a day but only 50% of households own radios. Rural women are always at work or at night tired asleep or tending to child, husband or sick household members. In a survey in Moyo only about half the women had any opportunity to listen and only 11% listens daily. Radio transmission is weak in remote rural Uganda which is where the most vulnerable women live. Also batteries are expensive and men may not be willing to allow their wives to listen to "liberating" programmes aimed at women (Nalwanga-Ssebina, 1988 in FHHBS). Less than 1% of households have TV sets and most broadcasting is in English. NGOs serve the rural areas with newsletters containing useful and interesting information some of it aimed at women.

Political deprivation: Within the home women are meant to be obedient to men and responsible for all domestic work. In all the activities men make all the decisions. "Female physical security is often violated through a wide range of abuse which includes rape, assault, beating, and female circumcision." ActionAid Uganda, (1994 p8) The degree varies between cultures.

Employed women are the least vulnerable, but they are subject to exploitation by male opportunists, sexual harassment at workplaces and often they have to leave their children at home unattended. Due to home demands they often cannot get promotion. Most employed women are in junior secretarial, clerk, teaching and other lowly groups. The 1988 census of civil servants showed that only 5% were women (UNICEF, 1994 p9)

Female-headed households are an emerging phenomenon. For example the mass recruitment of able-bodied men in Kabale to go to work on tea estates has produced a large number of female-headed households in Kabale. Women become heads of households due to male migration, separation or divorce, being widowed. or being a polygamous wife living in a separate house. Single mothers form a sub-group; no-one is responsible for her requirements, she has no control over resources, and is often subject to public scorn. A recent trend is that these women are regarded as a potential source of AIDS and looked on with community suspicion. In rural communities of West Nile this group is denied residence. Most societies have stern punishments to limit premarital pregnancy. As family ties loosen this group is likely to get bigger. The loss of labour and income of one adult lowers the economic status of these households. Women have less education, fewer legal rights and less access to support services and they can become entrenched in poverty.

Urban women: The role and place of the wife in the average urban family does not differ much from that of the rural wife. The urban community's sense of obligation to widows is questionable and not guaranteed by tradition.

Children: Almost half the population is under 15; the Department of Probation and Social Welfare estimates that 4 million children (about half) are vulnerable and live in difficult circumstances. The categories of children at risk include orphans (1.2-1.5 million), disabled children (approx 800,000), displaced or refugee children (unknown but possibly 300,000-400,000) and street children (2-4000) (UNICEF, 1994 p72) A recent article in New Vision claimed there were currently 10,000 street children. The infant mortality rate was higher in 1991 than it had been in 1968: 122 deaths under 1 year old per 1000 live births, compared with 120 in 1968.

Child malnutrition was seen as a major problem for children under 5 by the district leaders interviewed by UNICEF (1994, p46). The Uganda Health and Demographic Survey found that almost 1 out of 2 children is stunted (46.3% in rural areas and 25.3% in urban areas), and 1 in 4 wasted.

The New Vision studied 40 households in Kampala and found that half of them were sending their children to do work like hawking, fetching water, or manual work to earn money to supplement family income.

Adolescents (aged between 10 and 19) have particular vulnerabilities: to school drop- out, adolescent pregnancy, HIV/AIDS, unemployment, and separation from family."Girls get involved in love affairs with sugar daddies, not because they enjoy misbehaving, but from lack of financial support." (Kampala focus group). Seeley noticed the risky behaviour of adolescent girls who live with parents and want dresses and cosmetics but have no access to household cash (even though they work on their parents' farms) They seek lovers for fancy goods. Adolescents who get pregnant are seen as outcasts and can cause family breakdown. A Moyo focus group reported that a pregnant adolescent, even if married, becomes isolated at home, from school, and from her spouse. About 50% of rural teenage mothers whether they are married or not do not get any support from relatives, friends, or spouses (UNICEF, 1994 p44). Adolescent mothers have a high risk of premature births and low birthweight babies leading to health problems for the child.

Street children are under the age of 18, mostly live on the street and have no care and protection. The NGO forum for street youth estimates a total of 10,000 street youths living on and off the streets in Uganda (NV April 6 1994) In Kampala the number is estimated at 500. There are many street children in Mbale and Lira who have come from Kumi, Soroti and Karamoja. There are a number of reasons why children go on the street. Some are orphans, others come from families where the woman has divorced and wants to remarry but the new husband does not want the children. If they can afford it women will put their children in a boarding school or send to relative but often they are left to fend for themselves. Children may also be sent on streets to earn a living (Growing Out Of Poverty, p18); a small survey of the children found that only 37% were completely on their own and slept on the streets.

UNICEF believes that the problems experienced by women and children have been persistent despite economic recovery trends, increased social sector funding, rehabilitation of services and the return of peace.

Old people, alone or with dependants, are disproportionately at the bottom of the socioeconomic ladder. The AIDS pandemic means that many old have no middleaged children to care for them and they have to care for orphaned grandchildren.

Widows: Some widows are more vulnerable than others; some were the driving force in the household before the husband's death and cope well. Some widows are well provided-for by their husbands either before they die or with a will. They are at risk from opportunistic men. Normally the husband represents the family in the community and his death exposes his widow to problems she has little experience of dealing with. Typically she has played no role in marketing cash crops and is not familiar with coping with buyers, prices, and marketing, making her vulnerable to exploitation. Very few wives know the extent of the family's fortune, indebtedness, or major sources of income. This is not the legitimate concern of a wife. A rural widow's future life usually lies in the hands of the husband's relatives or sons. If the husband died intestate the relatives decide what she will get and whether she can stay in the matrimonial home. This will depend on the generosity and sense of fairness of relatives. She usually gets the home and some land the amount depending on how many children she has to bring up. She is often too poor to pay school fees.

AIDS widows are the most psychologically vulnerable as she too may be terminally infected. In the early days of the pandemic AIDS widows were so stigmatised as to be almost outcast; people's attitudes now are more sympathetic and understanding. Some, mostly those without children, leave their home areas to escape the stigma. If they have children one or more may be AIDS-infected. Community support in the areas most heavily infected is strong at an emotional and moral level but the ability to offer material support is weak. NGOs usually try to strengthen the community rather than target AIDS widows.

Orphans: It is taken almost for granted in Uganda that a father is not only the main provider for the family but also the motor of parental guidance and family stability; a child may not be called an "orphan" if the father is still alive. Loss of the father leads to physical deprivation, decrease in income, often an end to education especially for girls, vulnerability to social and legal exploitation which might include child labour, denial of access to the dead parent's assets, sexual abuse, and an emotional vacuum

The 1991 census found 784,000 orphans; 10% of children below the age of 15. The number of orphans due to AIDS is expected to increase 5-fold in next 5 years (GOOP 17). There are 115,000 estimated AIDS orphans and in 5 years time as many as 1 million children will have lost their parents due to AIDS. AIDs orphans place a large burden on the extended family and community as they tend to be related, and occur in communities where there are others in the same situation. They (and widows) are vulnerable to neglect, exploitation, and lack of resources for food, education, health services. Although the traditional extended family system is coping with the current burden of orphans there are emerging signs of failure.

Those who are fostered more likely to be neglected; for example they are less likely to be immunised (UNICEF, 1994 p27) They may also suffer discrimination involving lower access and attainment in schools. They may be used as a source of cheap labour (GOOP p38). The participants in UNICEF's study of districts officials said that selective denial of food to orphans and foster children existed (UNICEF, 1994 p48). "In my village people are selfish. You find a family of many kids and uncles who are rich. After the death of the parents, the uncles don't care for the children. The kids stop wherever they are in education." (Rukungiri)

In April 1994 New Vision claimed that Uganda is operating an archaic Adoption of Children Act dating from 1943. Many people want to adopt children in Uganda but the laws stand as impediments in their way. 1992 probation office research indicated there are about 75 children's homes most of whom were undertaking long-term care.

People with AIDS: AIDS is causing enormous suffering and becoming a significant contributory factor to poverty; badly affected areas like Rakai and Masaka place a high demand on health services while the depleted labour force has reduced agricultural production. AIDS in Uganda was first described in Rakai district in 1982. By June 1993 Uganda had 2,314 AIDS cases per 1 million of population. HIV/AIDS is now the leading cause of death for adults in Uganda and ranks 6th among childhood deaths under 5. There is a strong consensus among HIV/AIDS agencies and researchers that the number of cases is grossly under-reported and there are probably up to 5 to 7 times more cases. The principal explanation is that the disease, the patient and the family is stigmatized.

Virtually every district in Uganda has been affected by the spread of the HIV virus. The Uganda AIDS Commission (1994) estimated that over 1.5 million Ugandans are HIV positive. If preventive activities are not increased and attitudes and practices do not change it is expected that 3.8m people will be infected by 1999. The prevalence rate varies from about 2% in remote districts to 35% in urban centres.

It is estimated that in 5 years time 340,000 PWAs will need medical care. This exceeds the Ministry of Health's budgetary capacity treatment costs for 1 AIDS patient is equivalent to treatment of 70 episodes of malaria (ActionAid, 1994 p20). Hospitals and health centres have become primary health care units for a large number of AIDS patients; care is also often sought from NGOs, including mission hospitals and traditional healers. It has been estimated that the non-governmental sector provides nearly 2/3 health care in Uganda.

District personnel interviewed by UNICEF identified those most susceptible to acquiring HIV/AIDS as women, and adolescent girls as a result of sexual practices, cultural norms, and social expectations contribute to high risk behaviour. The linkage between alcohol and high risk sexual behaviour needs addressing. The fact that AIDS is stigmatized affects willingness to seek help and quality of support available.

People at risk include the well-educated professional classes, prostitutes, truck drivers, bar girls, soldiers. There has been a tendency to associate HIV's mode of transmission with risk groups rather than risk behaviours because of the difficulties of observing individual behaviour and groups provide a way of targeting interventions. However, in Uganda, HIV infection has spread well beyond the identified high-risk occupations and it is necessary to look at the circumstances under which people may be at risk. Poverty leads to poor access to risk-reduction information and less support for change to safer behaviour. However, HIV infection may also lead to poverty. In a study of 15 villages by the Medical Research Council (UK) Programme on Aids in Uganda covering 99,950 people in Masaka district, it was found that both male and female heads of the poorest households were most likely to be HIV positive.

In a related study of a cohort of 10,000 people in 15 villages in those aged 13-44 those HIV-positive were 60 times more likely to die than those HIV-negative. "The excess annual death rate associated with HIV-I infection was 5.3% per 1000 and in adults 7.9% per 1000; the highest excess mortality was 16.9 per 1000 in the age group 25 - 34." Half of all adult deaths were attributable to HIV-I infection; in the age group 25-34 this was nearly 90%. This is in a rural area where overall HIV-I adult prevalence rate is below 10%, a rate lower than in many other parts of East Africa. (MRC/ODA/UVRI Programme Annual Report 1993).

Women are at risk as individuals, mothers and carers - the low status and powerlessness of African women contributes to their vulnerability to HIV infection. A study of women in Kampala found that income was a good predictor of HIV status: the higher the income the less likely the infection. Two-thirds of the women said that women have boyfriends for economic reasons. In the Masaka area studied by the MRPA women without husbands who make their living from petty trade, hiring out their labour, and beer brewing or selling are called nakyeyombekedde ("I built my own house"). "The term is also considered by local people to be synonymous with prostitute" (Seeley et al, 1994)

There does not seem to have been a large change in behaviour. "The use of contraceptives, including condoms, is highest among educated adolescents (15-18%, about 4 times the national average)" (UNICEF, 1994 p45). One reason proposed for this failure is that sex is often part of an exchange relationship, for example it is exchanged for food.

People with disabilities: ActionAid Uganda estimates there are at least half a million people with disabilities most of whom are dependants and number among the very poorest. They do not have adequate access to services and treatment and find it difficult to get a means of livelihood and suffer from social stigma (as do their families). Growing out of Poverty estimated that 15-17% or the population was disabled, and that excluded the emotionally disabled (Growing out of Poverty, p19) while ActionAid quotes figures ranging between 3% and 11% of the population. In the 1991 census of children in primary school 1.5% were classified as disabled. Most of the disabled are dependants living in poor households. Many disabilities are acquired during childhood but survival beyond 19 is rare and most die before 45 from neglect or poor care. Disabled children from poorer households are less likely to survive during infancy. While many have confidence in their abilities society does not give them the opportunity to realize their potential. "Disabled children are not seen as human beings; they are isolated at home and not sent to school." (Kabale) (UNICEF, 1994 p21). People in several parts of Uganda believe that disability has a supernatural origin; there is a curse, probably on the whole family. They shun the disabled and families hide them.

School attendance by disabled children is limited by stigmatization, negative attitudes, and misconceptions of their physical and mental abilities. District personnel said it also affected attitudes to the children of disabled parents. (UNICEF, 1994, p21). Among the consequences of the civil war was a breakdown in primary health care, including community maternity services and immunization leading to the birth of many children predisposed to disabilities such as brain damage, poliomyelitis, and deafness. Also people suffered physical injury in the war and AIDS has contributed to neurological disorders. The disabled are often ignored in the targeting of essential services such as medical care and education. Even when they are targeted it is often difficult for them to access the services.

Disaster victims: Uganda is prone to earthquakes and pockets of famine. The waterhyacinth taking over the lakes is a slower kind of disaster which is particularly affecting poor fishermen who cannot afford engines to move to free parts of the lake, and whose nets get broken by the water hyacinth (see box).

The Western region of Uganda will always be vulnerable to earthquakes and the epicentre of the latest one to hit (6 February 1994) was at Kizomoro in Toro. Other districts affected were Kasese, Bundibugyo, Mbarara, Bushenyi and Kabale (New Vision, 28 February 1994). 500 dwellings were destroyed and the damage to public buildings and private property was estimated at US$19m.

While there has been dispute about the number of people who died in the recent famine in Kumi and Soroti, and a Commission of Enquiry is currently investigating what actually occurred, it is clear that many people suffered during the drought. "The few able villagers trek through thorny bushes hunting for wild fruits, tubers, green vegetables (ecadoi). Those who live near Lakes Kyoga and Bisina shores wade through the marshy waters to look for water lily tubers (ikorom) which have to be cooked meticulously because they are semi-poisonous." NV May 27 p16.

Refugees: According to the recent ActionAid report there are currently almost half a million refugees in Uganda. Recent estimates are as large as 1 million. In the late 1970s and early 1980s there were large waves of outmigration but by 1989 pacification and resettlement programmes had encouraged the return of 300,000 externally displaced Ugandan refugees. The intensification of the war in Sudan in 1988 and Rwandan unrest increased immigration into Uganda, although many of the Rwandans have now left. It is estimated that in the last 10 years over a quarter of a million people have taken refuge in Uganda; from Rwanda, Sudan, and fewer from Zaire, and Somalia. These periodic movements have stretched the meagre support networks in Uganda as the refugees tend to gather in parts of the country least suitable to handle the influx; Zairean refugees in camps around Arua and Bundibugyo districts; Somalis mainly in capital.

Refugees suffer from all the different dimensions of deprivation. They are subjected to indignities, humiliation and an uncertain future as well as other hazards, poverty included. The state of being stateless makes the refugee poor as there is often no access to resources such as land and capital. Since the draft Constitution has backdated Ugandan citizenship to 1962 this has increased the psychological difficulties and problems of integration of refugees.

Most refugees are children (often orphans), women and the elderly. They are exposed to a number of diseases as a result of poor sanitation in transit and settlement camps. Other problems include shortages of food and high levels of malnutrition, poor access to resources, inadequate educational funds; lack of accommodation and clothing. Many of these refugees have been in Uganda for years and no political solution has been proposed. NGOs are coping with burden of trying to provide for these groups.

Displaced people: There are an indeterminate number of displaced people from many causes including evictions and civil conflict. They contain a high proportion of women, old, and children and usually live in camps characterised by squalor, poor sanitation, shelter and nutrition. Their numbers greatly exceed the capacity of national institutions to cope (ActionAid, 1994). They move in large numbers usually with little prior notice and during the transition face a number of problems which make them vulnerable to poverty. The most recent examples are the Masindi Port Settlement (about 140 families displaced by war in the north), the Mburo Settlement in Mbarara district with 700 families displaced from Luwero during the war and Bugangaizi where 14,000 are now living in 21 settlements and 3 camps. They were evicted from Kabarole District during the war and hoped to return at the end of it but didn't. Others moved because of political persecution (eg from Acholi) while others have been evicted from forest and game reserves and land earmarked for government development. They have food scarcity, poor sanitary conditions, overcrowding leading to poor drainage, poor refuse disposal, low levels of income or none at all, poor or no social infrastructure (schools, health centres; markets) and limited availability of land leading to overcrowding and limited production.

The war-affected areas have suffered widespread physical destruction and severe economic dislocation causing losses in agricultural production and livestock, loss of parents, spouses, and children, and disablement and psychological traumas. Most areas of the country suffered during the civil unrest, although areas such as Gulu, Arua, and Luwero were particularly ravaged. About 500,000 people died in the various conflicts and many more were tortured, raped or maimed (ActionAid, 1994). The war-affected areas include Arua, Gulu, Rakai, Luwero, Soroti, Pallisa, Kumi and Kitgum affecting a combined population of 4 million people between 1979 and 1986. Rebel activities in the north have impeded the process of rehabilitation.

In Arua with a population of about 638,000, every part of the district suffered some kind of damage; 300,000 people went into exile in Zaire and the Sudan; postal services, police stations, communications network were all destroyed and most roads and bridges were damaged. In Aringa and Koboko counties three quarters of the schools were destroyed. There was a major decline in the production of food and cash crops: cotton cultivation stopped while the cattle population dropped from more than 200,000 before war to 120,000 in 1991/2. Before the war 22 cattle dips functioned; after the war only 2.

In Gulu, with a population of 340,000, most people were displaced from their homes, people were killed, roads and bridges damaged, government buildings destroyed. Education and health services collapsed; many children were orphaned. World Vision registered 70,000 orphans in 7 sub-counties in 1990.

The retrenched: the forced redundancies did not cause social unrest because of the generous compensation. However, it is likely that many of the retrenched will remain unemployed for a long time and may need programmes to help them re-enter employment. There is anecdotal evidence of numbers of retrenched people committing suicide.

Looking into the future is always dangerous but there are a number of factors which are important for our vulnerable groups which we can be fairly sure about and these are that the population will continue to increase, that the impact of HIV/AIDS will get worse, that crises due to climatic conditions, pests, and earthquakes can be expected at times, and that the risk of environmental degradation is high.

Uganda (85 people/sq km) is more densely populated than Kenya (43/sq km) and Tanzania (29/sq km). In 1991 there were 16.6 million. At the present rate of growth of 2.5% Uganda can expect a population doubling time of about 28 years. This will be influenced by changes in fertility rates and AIDS-related mortality

The vulnerability of resource poor farmers is on the increase (ActionAid Uganda RPF, p21). The increases in population are not being matched by improvements in technology, accessibility to services is limited and land is becoming increasingly scarce due to unfavourable tenure systems, population pressures and exploitation of poor by rich.

Predictions have been made about the impact of AIDS on Uganda's economy (Armstrong and Ainsworth, 1991) and include a decrease in agricultural production and a resultant decrease in foreign exchange. There is an estimate that by 1996 there will be half a million AIDS landless. Also life expectancy, now in the low 40s is predicted to decrease in the next decade, primarily as a result of to HIV/AIDS. It is estimated it will be 40.7 by the end of the 1990s; one of the worst in the world.

In Figure 2.5 we make some guesses about changes in the near future in the size of the vulnerable categories we have described. The numbers in poverty is only one criterion of the severity of a problem and it is important for policymakers and donors to consider the severity of the conditions of different groups at different times.

Poor people are well aware of the dimensions of vulnerability discussed in the paper - seasonality, life cycles, internal and external shocks - and plan their livelihood strategies with them in mind. Coping strategies can be adopted by individuals, households, communities, ethnic groups or whole nations. Here we are concentrating on individual and household strategies. Strategies can be seen in terms of (1) how individuals manage their poverty, and (2) the institutional forms and rules that they develop as they interact. Before turning to particular strategies adopted by the groups we have identified we will discuss general strategies that have been adopted by many poor individuals or households.

Selling services: People sell services; for example their labour or sexual services. In many parts of rural Uganda working for wages for others is despised making people reluctant to do it. In times of famine or similar difficulty distress selling of labour by females is becoming more common. Women in need may sell sexual services; this is sometimes accompanied by beer brewing and sale. The increased risk of HIV infection of the poor may be due in part to the income-generating strategies they adopt to survive. (Seeley et al 1994).

Asset-sales: Another general coping strategy is the accumulation of assets which can be used but also can be sold if necessary. This is an accepted way of smoothing consumption and saving at all levels of income; it becomes problematic for the poor when they are forced to sell factors of production, such as land or hoes. Another example is the selling of daughters for bridewealth, which often takes the form of cattle. Poor people sometimes have to sell the roof of the house, which is a sign of the failure of a man.

Charity: One source of aid is the churches who make welfare payments and NGOs who offer access to new income generating strategies. People also resort to begging and borrowing. Some borrowing is done on the understanding that repayment will be made when and if the person is able to.

In a major crisis people adopt a range of strategies, not necessarily in a particular order, although they are likely first to use strategies which do not affect future production prospects or fundamental social status (like selling the roof). "Peasants in the villages live from hand to mouth - in a famine situation which might last for up to 7 months they soon exhaust all they have in their gardens; then they eat any dry rations they may have in their granaries; as the famine progresses they resort to selling their labour to any rich peasant farmers in the village in exchange for food; then they resort to selling whatever assets they have at a giveaway price. In the eventuality everyone has no food the shops may have food but no-one has the money to buy it. Hunger slowly emaciates them and the weak ones die off. Traders are reluctant to take food to famine areas as people are so desperate they attack them physically and steal the food" (New Vision, 1994).

People in different vulnerable categories choose different portfolios of strategies to survive and some examples are described in what follows. For example the rural resource poor in Mitiyana sell labour to better off farmers and hire land for a limited time (usually 2 years) paying back a portion of crop harvested. They clear the land and plough it. The landlord repossesses after the hard labour has been done; 2 years is not long enough to recover production costs and make a profit.

There is anecdotal evidence from a number of places that some rural poor are coping by marrying their daughters earlier for the bride price. A falling age of marriage may be a sign of a long term slide into poverty.

During the drought in Masindi between December 1991 and June 1992 there was widespread stealing of chickens, goats and cassava. The poor also begged for work for cassava from capitalists and rich peasants who had food in store. The selling of land in a crisis leads to increasing social differentiation.

The landless have limited options; their poverty arises from the lack of non-agrarian employment opportunities. Rural labour is poorly paid and demand is irregular. The rural landless have little to equip them for skilled or urban employment. They are frequently unable to obtain suitable housing, health care, and education. More and more are moving to urban areas, increasing the numbers of the urban poor. They may become rural labourers on tea or sugar plantations.

Women normally cope with their problems by marriage and submission to men. They also engage in petty trade, the sale of labour, sexual favours. Joining extended families is sometimes difficult as they find themselves unwelcome in male-headed households. Women's groups, co-operatives and associations have been one of the major survival strategies, especially for raising incomes. Widow inheritance, common in some tribes, is often embraced by women as a security to land and other factors of production.

During the 1991/2 drought rising prices led to the distress sales of women's labour at lower than the village norm (UNICEF Situation Analysis, 1994 p29). Urban widows are usually employed anyway and options exist including retailing agricultural produce, brewing and selling alcohol, preparing and selling cooked foods, and prostitution.

For the urban poor temporary employment bringing in low wages is common, as is petty trade, hawking, prostitution, casual labour, local beer selling, and housekeeping. Street children in Kampala "congregate on specific streets; they sleep in tunnels, grounded vehicles, abandoned railway wagons, market stalls, underneath garbage cans. They eat left overs picked from garbage cans or buy cheap food from street vendors. Some cook in dirty tins picked from garbage." They are often ill with stomach pains, cough and colds etc and have sores and wounds mainly caused by car accidents and fighting. Sexually transmitted diseases are evident. A recent study by Friends of children suggested that Baganda and Rwandese refugees were the dominant ethnic groups.

In Jinja, Busia, and Mbale migrant tribes from drought-stricken Karamoja are dominant. Boys outnumber girls by 9 to 1. Many of the girls are smart and appear in the evening only. Many have contacts with relatives and even parents. The study by FOCA in 1992-3 showed several drugs being used - opium, alcohol, tobacco, khat, petrol. A typical day involves waking at 5.00 and starting by lifting garbage at flats; others fetch water or carry edibles in the market. At about 9.00 they congregate around garbage cans to pick out valuables; tins, rubber bands, bottles, and food, some of which is sold in the slums. Some do manual work for pay.

War-affected people are developing ways to cope with stress from civil strife. In Luwero and the north and northeast many families and communities were so disrupted that they ceased to exist as coherent units. In the Luwero triangle many village communities have formed organizations known as Munno Mukabi to offer help to war-disadvantaged including orphans.

In Masaka and Rakai there are some households headed by child orphans; by the time the AIDS victim dies the family is drained emotionally and financially. Young and poor orphans have fewer coping strategies; the majority depend on already established gardens and exchanging their labour for a little money and food. In urban areas orphans engage in hawking, petty trade, housekeeping, restaurant attending, and begging.

There is a strong feeling in the country against orphanages and it is Uganda government policy to avoid them as much as possible. However the wardens see them as temporary refuges. The government regulations are fierce and seem designed to make life difficult for the orphanages which play a life-saving role in caring for unwanted children and need help.

Household surveys measure poverty at a point in time but while some people are firmly stuck in poverty over a longer-period others are moving in and out of it over shorter or longer periods. People can fall into poverty as a result of a shock, or series of shocks to the household, or they can sink more gently into it. Once in they may recover fairly quickly or may find it difficult, if not impossible, to recover. One of the reasons for this is the interlinking of different aspects of deprivation. In this section we examine some of the processes which need to be understood if a dynamic view of poverty is to be taken.

Shocks are events which either occur naturally, or as the result of ongoing activities external to the economy, community, household or individual. They are not under the victims' control and may be damaging or beneficial. In either case they may propel the individual or household into a poverty or wealth cycle. Natural shocks include rapid onset shocks such as earthquakes and floods which cause crises that are usually temporary (although they may have long-run effects); the earthquake in Hoima in February 1994 is an example. Another form of shock is the slower onset emergency such as the famine in Kumi and Soroti in the middle of 1994 and the spread of water hyacinth in Lake Victoria.

International and local economic shocks such as sudden changes in input and output prices have profound impacts on households, particularly rural ones. The increase in the coffee price is a benign shock. Many households in Uganda have also suffered or are at risk of suffering from internal national or community level shocks, such as civil war, banditry, cattle-raiding, epidemics, evictions from areas gazetted as national parks, changes in government policy. Recent or current examples include the Kony rebellion, the Karimojong raids, and the cassava mosaic epidemic. The five changes of government between 1980 and 1986 made many bureaucrats vulnerable to imprisonment and poverty.

Some areas, particularly in the north and east, are at risk of multiple shocks as the following example shows. "Life has been disrupted in the remote small mud-and-wattle grass thatched village of Adugugudu Apasakamerere in Ongino sub-county, Kumi district, since late last year when the drought swept through the area like a wildfire. The drought came after the Karimojong rustlers in the late 1980s with ruthless efficiency grabbed most of the cattle formerly used for ploughing. Diseases and malnutrition completed the onslaught. "( New Vision, May 27 1994). There was a general drought in 1991/2, while in 1994 the drought in Kumi and Soroti was compounded by cassava mosaic, baboons and warthogs. District officials attributed the famine in Kumi and Soroti to the 1993 crop failure, the influx of Karimojong herdsmen and their families early this year in search of pastures, water and food, big sale of the 1992 sweet potato boom instead of preserving some in the traditional way in anticipation of food shortage, and long drought. They also mentioned the ploughing handicap. The Iteso lost their cattle in the cattle rustling during the conflict in the area in 1987.

Households are at risk from individual-level shocks, such as an episode of illness, a death, divorce which can project them into a poverty cycle. The female-headed household interviewed in Kamuli (see Box) is one such example.

Longer-term uncertainty and variability are features of an unpredictable economic and political environment. For example, if prices of cash or food crops fluctuate wildly from year to year, growers are unable to plan in a rational way and some are likely to enter the poverty cycle as a result. They are also likely to choose safer, lower risk activities with a high probability of mild poverty than higher-risk/higher-return activities which might lead to acute poverty.

Areas or households which suffer repeated shocks get more and more vulnerable as time passes. "In recent years Uganda, like many other countries in the sub-region, has experienced famine repeatedly. This has led some districts to have drastic reductions in food production, leading to acute food insecurity." (P Nkono-Luganda, New Vision May 27 1994)

Vulnerability. There are people who, while they are not necessarily poor all of the time and in all of the ways suggested, are vulnerable to it. They do not have means to fall back on if a crisis occurs and these are the subject of Chapter 2. Others fall into poverty at specific times - eg in the hungry season or when school fees are due. Some of these people may be relatively well-off at times but live under the constant threat of losing everything. Pastoralists threatened by drought are one example; squatters liable to eviction are another.

Growing out of Poverty (p11) claimed that hospital admission records and field interviews suggest that malnutrition increases dramatically in the preharvest season. It is possible that many households fall below the poverty line for three months of the year. While families may sell food during the harvest season the same families have to buy it in later seasons when prices are considerably higher; inflation and intra-year vulnerability are seasonal (ibid p54) . A PRA in Mbale found that the agricultural workload peaked at a time of food shortage, low cash income, and high exposure to illness (Goetz et al, 1994). In Karamoja the nutritional situation deteriorates during dry season as young adults migrate to more fertile lands with cattle leaving behind the weak - women, with young children, the sick and the elderly. Recurrent droughts plus insecurity are stopping herdsmen bringing the cattle back (GOOP p 24).

Cycles of poverty and wealth: People engage in economic activity through time and the cycle of investment, production, consumption and accumulation to be followed once more by investment can take place at different levels of wealth or poverty. An adverse or beneficial shock at any point in the system will have repercussions for later cycles. People can also get into spirals - heading either up or down. Government policies should be concerned to establish what projects people down to lower cycles and try to provide safety measures against this, and also how people climb up to higher cycles and facilitate this.

It is possible to identify other types of vicious cycle. One example is the malnutrition-diarrhoea cycle which babies get into; in Mbarara a study found that malnourished children have a significantly higher risk of death from malaria, measles, diarrhoea and acute respiratory infections. Also childhood infections, especially diarrhoea, acute respiratory infections, and measles, predisposed children to malnutrition; loss of appetite leads to decreased intake which leads to decreased absorption. Conversely deworming can lead to decreased wasting. (UNICEF Situation Analysis. 1994 p49). A second example relates to environmental degradation: poverty related to land shortage leads to over-cultivation which leads to environmental degradation which leads to lower yields which leads to greater poverty.

In some African countries a coup-corruption cycles gets established. Favoured bureaucrats who may have misused office fear Commissions of Inquiry under the next regime. Accordingly they use corrupt practices to store up funds in case of another coup.

In the normal course of events throughout their life cycles people become richer and poorer depending on their circumstances. Successful development involves a general raising of standards of living but within that some people will become considerably richer and others will become poorer. In many cases the wealth of some depends on the maintenance or creation of the poverty of others. The "green revolution" in Asia has raised measures of GNP but it has also increased inequality, and arguably poverty for those who lost their land. It is important to examine the reasons why some groups are able to move out of poverty, while others remain there or become even more badly-off, focusing in particular on the extent to which, and conditions under which, growth for some people involves increasing poverty for others.

There is case study evidence from some parts of Uganda suggesting that growth, particularly in non-traditional food crops such as maize, may be accompanied by increasing rural differentiation and some people argue that, in some areas trade liberalisation, the move into food crops for sale and export, and the lack of improvements in productivity, are increasing "land-grabbing", wage labour and hence poverty. Food shortages, particularly during droughts, force the poor to sell labour to rich peasants. The more they face crises the more they become dependent on wage labour and while the farms of the rich are being attended to the farms of the poor are going to bush. (Bazaara: 1992 study of Mpumwe village in Masindi). A precedent for this type of behaviour can be found. The 1960s witnessed substantial evictions of tenants by landlords who wanted to occupy their land (ActionAid, Uganda, p8). The provisions for protection of customary tenure were overridden, particularly in the north, where people could use political influence to acquire land, and in Buganda. The rise in land values provided an incentive.

ActionAid Uganda, on the basis of a PRA exercise in Mityana, Kumi, Arua, Kibaale, and Bushenyi Districts suggests that it is reasonable to speak of 3 types of peasant: rich; middle and poor:

"The Rich Peasants have large tracts of land where they practice large scale farming (sometimes mechanised), hire labour, can afford using chemical inputs and usually do not get directly involved in the day-to-day farm activities. They sell in bulk and realise large profits. They have regular surpluses above the needs of simple production and are able to combine the returns from labour with those of small scale exploitation such as renting out land or hiring wage labour."

"The Middle Peasants have limited access to the factors of production. They normally rely on family labour but can also afford hiring labour. The volume of production is less than that of the rich peasants and they participate in the day-to-day running of the farm activities and have full control of the labour processes. ... The agricultural tools used are simple and are owned by the family. The hand hoe and panga are a common phenomenon."

"The Poor Peasants are those farmers who have very limited access to a combination of factors of production. The resource poor peasant farmer lives in a vicious circle of poverty. Whatever is produced is simply for subsistence and sometimes the harvest is sold to meet basic home requirements. .. He cannot afford to rent implements of labour or land. This results in the poor peasant to enter into relationships with the middle and/or rich peasant for survival. These relationships normally result in exploitation." (ActionAid Uganda REF p3)

In Masindi Bazaara distinguished between capitalists and wage labourers and three types of peasant: rich, middle and poor. The capitalists gain income from diverse sources including agriculture, shops in Kigumba, maize mills, pit-sawing and hiring out ox-plough and tractors. The wage labourers get most of their income from the sale of labour to rich peasants and capitalists, and sometimes middle peasantry. They have no access to land or instruments of production except as "borrowers" (they have to give chickens, beer, agricultural produce and labour whenever it is required). Rich peasants operate partly with family labour and partly with wage labour and labour by poor peasantry. Middle peasants mostly use household labour and communal labour and supplement their income by brewing waragi and kwete. Poor peasants survive by the partial sale of labour to capitalists and rich peasants and family labour. The capitalists and rich peasants have enough resources to act as an effective force in the maize market. During the harvest period prices fall and the poor, who are desperate for cash, sell at low prices, usually to the rich. They store the maize and sell it when there is scarcity, or they try to act as commission agents for long distance produce dealers. When the poor suffer acute health problems they sometimes resort to selling pieces of land.

Advantages and deprivations are handed down through the generations. For example poor parenting may be associated with lack of access to education which in turn exposes adolescents to multiple economic hardships when they can't find employment. An HIV+ mother has a 30-40% chance of transmitting the virus to her child. " Of non-infected children born 18% will be orphaned or die of causes which could have been prevented if mothers had been healthy." (UNICEF, 1994 p63). Children born to women with secondary or higher level of education have half the risk of dying under age 5 compared with children of women with no education (ibid, p24). Many of the current political elite are the children of a previous elite.

The fact that many adults raising families today are suffering from post-trauma psychological problems puts stress on families which can produce inadequate parenting, including lack of food, shelter, love, nurturing, and protection (UNICEF 1994, p53).

The five dimensions of deprivation tend to be highly correlated, as are their opposites. The poor are often trapped in a cycle of poverty and cannot break out of it without external assistance in changing one or more of the dimensions of deprivation. Rich people can convert their wealth into high status and political power. People use political office to increase their wealth and status. In the village in Masindi studied by Bazaara landowners and rich peasants were involved in a constellation of alliances and controlled politics at the village level which mainly took place in the "church associations" and RCs.

Land is a source of social prestige; its lack creates a social outcast group without confidence who cannot participate in the country's development (ActionAid Uganda:L. 1994 p10.) Cultural values which promote early marriage and pregnancy for girls result in multiple health risks because of physical immaturity and poor nutrition for children due to emotional immaturity. The personal experience can be compounded "School dropouts lack identity; they kind of get lost socially and economically. They fit with neither the illiterate nor with the educated people; the drop outs are only half-baked " (Nebbi UNICEF, 1994 p18). Nonenrollment in school and dropout were seen by district leaders to have far-reaching consequences for children and adolescents throughout adult life. Low literacy and numeracy and the lack of job skills in turn lead to the inability to read and write useful information, the inability to participate in public meetings, the inability to count money leading to being cheated while selling produce, the inability to take up certain jobs, and idleness. Idleness and lack of finance leads to indiscipline, delinquency, crime, early sexuality, early pregnancy, early marriage and higher HIV/AIDS risks. (UNICEF, 1994 p22).

Traditional beliefs are a significant determinant of malnutrition among children and to a lesser extent women and adolescents.

Public insecurity, for example war and cattle-raiding often lead to displacement and poverty. Then survival becomes more important than education. Natural disasters like drought and famine affect school attendance. In Kotido and Moroto where drought is a recurrent problem school attendance is strongly linked to availability of food at school (UNICEF, 1994 p20). Rustlers and drought have had a big impact on education in Kumi and Soroti The headmaster of one primary schools said emaciated half sleepy pupils can't concentrate. The teachers often have 1 meal every 2 days because salaries are often delayed. At Kanapa primary school out of 315 children enrolled at the beginning of the year only 185 still attended in May; many parents cannot pay the school fees. At Akide primary school out of enrolled 300 only 75 still come to school and Kabosi primary school only has 48 out of 200 enrolled pupils still attending.(New Vision May 27 1994 p16).