Zambia - Poverty assessment (Vol. 5 of 5) : Participatory poverty assessment

Report No. 1 2985-ZA Zambia Poverty Assessment (In Five Volumes) Volume V: Participatory Poverty Assessment November 30, 1994 Human Resources Division Southern Africa Department Africa Regional Office Document of the World Bank  ZAMBIA PARTICIPATORY POVERTY ASSESSMENT SYNnTESIS REPORT Report authors: ANDY NORTON DAN OWEN J.T. MILIMO Researh team: S. Chimula N. Bubala M. Nabanda v. Mbewe C. Njobvu F. Kondolo P. Ponga S. Chama C. Kalamwina PRA Trainer Meera Shah  ZAMBIA PARTICIPATORY POVERTY ASSESSMENT ACKNOWLEDGEMENTS AND PREPARATION TEAM The research on which this report is based was carried out by two teams from the Rural Development Studies Bureau of the University of Zambia, under the supervision of Dr J. Milimo. The team members were: messrs S.Chimuka, N.Bubala, M.Nabanda, V.Mbewe, C.Njobvu, F.Kondolo, P.Ponga, S.Chama, C. Kalamwina. The work was commissioned by the Southern Africa Department of the World Bank to input to the Zambia Poverty Assessment. Tle research took place in the period from September to November 1993. Mr Dan Owen (Southern Africa Department, World Bank) coordinated the exercise and participated in the fieldwork. Dr Andrew Norton (AFTHR, ODA secondment) participated in the training exercise, and was responsible for overall research design and the preparation of the Field Guide. The UK Overseas Development Administration supported the costs of an input of traimng in Participatory Rural Appraisal methods (Ms Meera Shah, IDS), and SIDA supported the teams' in-country costs. This report has been prepared by Messrs. Norton, Owen and Miimo following on from a synthesis workshop held with the research teams. Special acknowledgements are due to: the research teams for their enthusiasm and commitment; Mr Steen Jorgensen, Task Manager for the Zambia Poverty Assessment for his encouragement and support; Ms Meera Shah for the very high quality of her training input and a workshop report that was of considerable value in preparing the final report; and above all to the communities who participated in the research exercise for giving their time to share their views and analyses of poverty in Zambia. i  EXECUTIVE SUMMARY Rationale The Zambia Participatory Poverty Assessment is a component part of a series of studies leading up to the preparation of a poverty assessment for Zambia. The overall objectives of the poverty assessment are to establish a poverty profile for Zambia and to identify appropriate actions for poverty alleviation. The work will help guide Bank strategy and will provide input for revisions to the Bank's work program, on the basis of a Poverty Alleviation Action Plan, which will also seek to identify appropriate actions for other agencies including government, NGOs and other donors. The PPA contributes to this process by including within the analysis the views, perceptions, experience and preferences of the poor themselves. Objectives The objectives of the participatory poverty assessment are as follows: to explore local conceptions of poverty, vulnerability and relative well-being in poor urban and rural communities in Zambia * =Sto explore what the poor themselves see as the most effective actions for poverty reduction which can be takhn by i) individuals or families, ii) communities, iii) government agencies, iv) other ;nstitutions * to investigate what people in poor urban and rural communities see as the main concerns and problems in their lives at present and how these have changed over the last 5-10 years * to investigate local perceptions of key policy changes related to economic liberalisation Methods The methodology for the PPA included a mix of methods and techniques known to the teams under the labels 'Beneficiary Assessment' and 'Participatory Rural Appraisal' (some, for example, focus groups and semi-structured interview techniques, are common to both). These methods include: * unstructured and semi-structured interviewing * focus group interviews * participatory thematic mapping * wealth and well-being raning * institutional diagrams (venn diagrams) * matrix ranking and scoring * seasonality diagramming * trend analysis * livelihood analysis * time-lines ii  The fieldwork can be viewed in terms of two major methods. Firstly, an Interview Guide was prepared for the purposes of conducting semi-structured interviews with individual informants and focus groups. The research teams were encouraged to use the SSI Guide flexibly, with researchers probing and seeking to follow relevant and interesting leads as they came up. Aside from the SSis the teams faced the challenge of using a range of highly participatory research methods which are designed to access not only the knowledge and information possessed by the rural and urban poor, but also their own analysis of their situatioi. There is a limit to which this type of method can be applied according to 'blueprints'. Teams were encouraged to think for themselves concerning how to investigate topics using sequences of methods - and mixing the 'visual' methods with the semi-structured interviewing. The ten research sites were selected in order to represent a variety of communities differentiated by: (i) rural/urban characteristics (ii) mode of livelihood (iii) cultural/ethnic group (iv) agro-ecological zone (v) level of access to infrastructure and services (vi) level of integration with markets. The main livelihood groups are farming, fishing and those in formal and informal employment. Farming groups are further classified by the predominant staple food crops under cultivation (maize, millet, sorghum and cassava). This distinction is reflected in the grouping of agro-ecological zones - tht is, the high rainfall areas are associated with cassava ad millet while the lower rainfall areas with maize as the staple crop. Farming also includes the agro- pastoral groups to whom cattle are extremely important. The principal ethnic/linguistic groups covered in this study are the Bemba, Lozi, Nyanja and Tonga. In terms of access to infrastructure the area which is the most isolated by far is the Western Region - which is frequently omitted from studies of this kind for just that reason. The selection of participants was largely effected through selecting poor urban and rural communities - selections based in part on the field teams' prior knowledge (some of the communities had been visited before in the course of other exercises) and in part on the advice of key informants. Within those communities a range of people were usually interviewed. Many of these people would not necessarily consider themselves 'poor' in local terms, being full and active members of their own societies - the great majority would come under a national poverty line, however this was defined. Even the wealthier members of these communMies suffer, as we shall see from a range of problems which affect everyone within a given community - poor roads and infirastructure, lack of access to safe water supply, lack of social infastructure and services, lack of access to markets. iii  bont9m PoErMsdeUUe In Zambla:, a Sl=M gf Concluion and Prioritlesfo ITe main conclusions of the Zambia PPA can be summarised as follows: K al Percepons of oerty Local conceptions of poverty reflect a variety of factors which are locally specific, and also have commonalities between different rural and urban communities. It is useful to distinguish between the characteristics of poverty which are seen as applying at the level of the householdlindividual, and those seen as applying at the level of the community. * Key dimensions of local understandings of rural poverty at the level of the householwdndvdual include: - assets (variable by production system - eg. cattle for stores of wealth, manure, traction etc.; fishing equipment; granaries etc) - aetivities (especially the necessity to engage in 'piecework' to obtain food during the lean season) - consumption (quantity/variety of food) - socdal status: the most striking feature to emerge from the wealth ranking aercises was the issue of gender. Spec#cally this referred to 'women without support' - the dimensions being gender, lack of adut children, widowhood/divorce/lack of a partner, age. Another issue of imWortance In distnguishing the core poor in local views was the question of diablit. Significant factors which affect livelihoods at the level of the rural community were found to include: - access to services and social infastructure - access to transport, productive and market infrastructure - the natural resource endowment of the community. Key dimensions of urban poverty at the level of the household/indidual include: the control ownership and use of assets and management of resources (assets include those necessary to maintain basic standards of participation in communities - basic essentials for cooking, washing and dressing; productive assets; land - especially legitimate tenure status in urban areas); consumption (especially in terms of quantiy and quality of food); 3Por a more detailed discussion and summary analysis, see sections 3 and 5 in the main text on Policy Issues. iv  - security of employment (petty trading, income diversification, and in some sites reliance on self-provisioning in agriculture emerged as means of adaption to rapid changes in the labour market); - soidal statu8 cross-cuting all the themes emerging In the discussions on poverty In the urban comnities were Issues of gender, age, disability and social ssfrom marriage - In wealth ranking exercises In urban, as rural, communites, routinely the lowest category would consist solely offemales without support (unmarried, widowed, divorced, without adult cidren stpporting them). * The foremost factors perceived to determine poverty at the level of the urban community include: - access to services - access to social and economic infrastructure - access to welfare and vocational ting - crime and prostitution - alcoholism Lv A&=es to Serfiew 1. social Servces HealWcare * ..Access to healthcare is a very high priority in poor rural and urban communities - this was the most consistent finding from the problem ranking exercises. * uIcreasing costs of healthcare were said to be increasingly restricting access of poorer households. Among the consequences listed in some field sites were an increasing reoourse to home treatment and traditional healers (although traditional healers tend to be seen as specialists in certain disease). * High cost of healthcare is a particular problem during the seasonal stress points when disease problems are rife in urban and rural areas due to the compound effects of the rains and other factors (poor nutrition and in urban areas poor sanitation). The period from October to April combines the seasonal labour peak in rural areas with a squeeze on income, and a high incidence of disease. * In most areas, and for most illnesses the first recourse is to home remedies. This trend was said to be increasing as a result of cost-recovery measures in the formal healthcare system. * In relation to rural clinics poor supply of drugs was observed consistently as a problem. v  * Observations that staff were rude and arrogant in relation to the poor were consistent for both hospitals and clinics. * The condition of the access road to the area during the rains was a major concern of most of the rural communities In terms of access to specialist care. * Declining quality of service was observed in poor urban areas, related to under staffing of facilities as well as patient congestion due to inadequate space, and lack of medication (drug supplies). Education * In rural areas access to education did not emerge as a major priority 'need' - access so primary education was felt to be adequate, although questions were raised in two communities about performance of teachers and supervision. In terms of the levei of demand for education, the role of the children within the farming system was a deteinant. - along with factors in the system of bridewealth and marriage (eg. Tonga communities where young boys have dry-season herding duties and parents fear that education will lower the 'value' in terms of bridewealth cattle of their daughters). The issue of rising costs of education was frequendy raised as a problem at the level of the household. This was related not only to school fees, but also to uniforms, books and materials. These costs arise at the point when household incomes are weak (the pre-harvest season) and when poor households with a food deficit are facing the problem of trying to obtain emergency food stocks to cover the period up to the harvest. : In urban communities people's perceptions were in the main limited to the primary school environments, and grievances focused on limitations of access to schooling, generally lack of places. In some cases, frustrations associated with constraints to access had been compounded by cholera epidemics, which have forced the temporary dosure of some schools. Dissadsfaction wish eachers'perfonnance was broadly shared. In an alarming number of cases, teachers had been reporting to work drnk - evidendy related to low morale and poor motivation. In others, some did not regularly conduct classes. In terms of quality of education, parents voiced reservations about contact time and the amount of time children were spending our of the classroom in the course of the school day, ofken doing physical exercise or manual labour. Poor facildes and lack of supplies and furniture - basic desks and chairs - were a subject of concern amongst parents. Schoolfees were cited as a significant barrier to access. Communicadon !,etween school administration and community was found to be lacking. Water and Sanitation In rural areas the distance, reliability, conditions of physical access and quality of the local water supply (especially the dry season water supply) was the other issue (alongside healthcare) that received a very high consistent priority in the communities - especially from women. In general women ranked this issue higher than men - in communities where the dry season water source is a long way from the village women routinely spend five hours every day fetching water. In addition, material on seasonality and health clearly demonstrated that in communities where dry season vi  water quality was a major problem this created a clear seasontl peak of dysentery, bilharzia and other water borne disaes. In communities wbere men were concerned about dry season water supply for catle this raised the issue of water supply gready as a priority among men - although the interests of the genders are different in terms of the type and purpose of water supply. * Perceptions of probhas associated with water and sanitation varied across the urban communities visited, ranging from general satisfaction to great dissatisfaction in another. In poorly provisioned compounds, inadequate infrastructure was seen to very cleady add to the labour burden on women, who often are compelled tw wait in line for hours, or to walk to neighbouring compounds to fetch water. People were discerning with respect to the health complications symptomatic of infrastructural deficiencies (pwoor water, sanitation and garbage disposal) and linked outbreaks of diseases such as cholera and dysentery to dirty water, infused with the dirt from latrines built close by. Agiulturat Extension and Research [RURAL AREAS] As is noted below, the principal trend observed by villagers in rural communities over the last two years has been the collapse of the government's system for maize marketing. The production of hybrid maize is no longer considered a worthwhile activity in most rural communities and people are switching to other crops - especially in order to gain cash income. As the government's extension system has traditionally been geared to promoting hybrid maize the decline of this system has accentuated a general perception of uselessness and irrelevance in relation to this service. These negative perceptions were particularly strong among women. In contrast to the general apathy concerning agricultural extension services, there was a considerable demand for agricultural credit - usually perceived in terms of access to supplies of fertiliser at 'affordable prices'. Uvestock services, especially vaccinations for catte and dip tank treatwents to prevent 'corridor dLaase' were very highly valued in Tonga areas. There were consistent complaints in one field site that the state's withdrawal from this area of activity had led to an outbreak of disease which had reduced local catfle holdings by over 50 per cent. Grndng mils [RURAL AREAS] Lack of access to grinding mills was raised as an issue in four out of six rural sites. Grinding mills are preferred over traditional and 'alternative' technologies (hand- powered mills) for the following reasons: they significanty reduce labour demands on women; they reduce wastage; they grind the meal finer than hand-operated mills. Wefare Cenres [URBAN AREAS] This was raised as an issue in all the urban sites, and its function was perceived on two dimensions. Firstly, as poverty was identified with lack of employment, people expressed a need for income-generation support. A welfare centre would facilitate acquisition of skills, catering for apprenticeships, and provide such services as trades vii  training and credit for small enterprises (perhaps in the fornm f ! revolving fund). Secondly, the provision of welfare was conceived in safety net terms, to cushion and provide 'social security' to the increasing number of socially and economically marginalised groups. These groups were identified as, for example, streetchildren in Lusaka, and the elderly, disabled and generally female headed households in Mongu. A welfare centre in the latter case would serve as a proxy for local social institutions for mutual support, the capacity of which to assist households in times of particular hardship has become seriously diminished. The articulation of need for social welfare institutions in the conununity, whether serviced by Government or other, can best be understood as an indication of general concern by community members with issues of long-term vulnerability. 2. Access to social and economic infrastructure Transport irastrcture Mobility emerged as a key concern in urban and rural areas. Infrastructural deficiencies bedeviled satisfactory provision of public transport and access to social services, especially education and health (particularly access to emergency curative health services). The lack of transport also emerged as being particularly resented in the context of social ceremony and customs - lazk of transport to ferry the dead was bemoaned in numerous cases. In urban areas, poorly maintained transport infrastructure was also linked to multiplying crime, as poor or absent street lighting made the compounds perilous at night, and poor maintenance of roads was identified as a significant health hazard in and of itself (sore eyes and coughs from the dust during the dry season). In rural areas, a prominent direct linkage observed by local people was with access to markets. An impassable rainy season road takes its toll on incomes in most of these areas as traders cease to buy produce. Where dry season vegetable farming or fishing were significant activities, people were aware that the level of incomes they could expect from these highly perishable products was dependent to a large degree on the condition of the road. In sum, crossiectoral benefits were recognised for roads by both rural and urban populations. Markets and trading Also relating to people's assessment of infrastructure in urban areas was access to and organisation of markets. In Mongu, community members, e3pecially traders, felt that the provision of market places would benefit the poor. A videly shared sentiment was that, at the very least, punishing people for operating from their homes in the absence of markets was wrong, and that a system of licensing should be instituted in its place. 3. Other Issues Access to Natural Resources [RURAL AREAS] In poor rural communities, access to natural resources forms the predominant element of sustaining their day-to-day livelihoods. Access to land for farming, and for grazing animals, comes to individuals by virtue of membership of social institutions of kinship viii  and community. In addition to directly productive uses of natural resources for agriculture and fishing, the following other imensions of rural livelihoods are also significant: access to foraged foodstuffs (particularly critical In stress periods, both seasonal and due to drought, and for poor households which regularly face food deficits); access to fuel; access to water (see above); access to herbs and roots used in traditional medical treatments; access to 'bushmeat'; access to other foraged materials which can be sold for cash income (thatching grass, other building materials). WoMlerzce [UAN AREASI The very real threat to prsonal physical security at the community level was attested to in each of the study areas. A picture of a spiralling cycle of crime and violence emerged from the testimonies in the compounds. Many people related this to drunkenness, and the late closing hours of bars in the compounds, but it also can be seen as evidence both of the desperate measures some people perceive as their only recourse in times of such persistent adversity and as evidence of a growing number of young people "turning against society". In all the areas visited, people lamented the breakdown of law and order, the lax civil code and the apparent failure on the part of the police force to pursue criminals, investigate cases, punish offenders or restore order. Lrdl tenvC Two types of insecurity relating to home and property emerged in discussions in the communities. Firstly, the question of legal rights over land and property arose as a serious concern (above all in the speculation as to whether customary land would be brought under the aegis of township councils as council land (and thus be at least entitled to council services)). On the other, the structural vulnerability of people's homes was highlighted and portrayed in many instances as an indicator of poverty. Se naliY and Welfa Seasonal food supply was found to be a regular problem for a significant number of households in most of the field sites in an average year. In two of the rural sites where there was least diversity of crops (and monocrop dependency on maize), this was seen as a major recurring problem facing most households. In those areas, crop diversity was seen as boosting resilience to shocks such as drought or crop disease, as well as mitigating the impact of seasonal 'troughs'. In urban areas, households are drawn through cycles of recurring resource deficits in which scarcity of food is the main feature. Seasonal Variation in Incomes and Expenditures: the striking point of consistency in the material gathered on this issue was the vulnerability associated with the lean months of November to January/February, where low income and high expenditure are at their peak. Characteristic of this season are: * high incidence of diseases; * fees for health and education; * borrowing from money lenders; ix  * very low income; * very little food available; * scarce fuel; * outgoings for uniforms and other school requirements; * investments in agricultural Inputs for maize production; * Christmas celebrations. Household vulnerability In urban areas appears to intensify significandy on another cyclical dimensions: on a monty basis - towards the end of each month, especially for salaried employees (but also for non salaried and Informal sector workers as general purchases and consumption are tightened) - as money becomes scarce and consumption Is economised. In all of the rural communities the period of peak labour demand in farming coincides with the low point for food stocks, and the peak point for malaria which is judged to be the most serious health hazard in most communities. Women, with a heavier workload than men, suffer more. A clear =d in rural livelihoods related to maize marketing, especially governmental policy and the history of govermment involvement in this sphere. There is a widespread perception (articulated in the five rural field sites where this was an issue) that the goverment is still controlling the producer price wbile liberalising the costs of inputs, and therefore market conditions have deteriorated to the point where it is scarcely worthwhile to grow maize. Among the consequences that were listed as a resuit of the problems with maize marketng were: i) swtching crops: away from growing hybrid maize for eith,r own or cash income; ii) barter systems.: an increasing resort to barter systems of exchange as a result of the increasing breakdown of the state run system of maize marketing was reported in one field site; iii) Increasing slgnkicance of offfann income-generating acthtes. In rural areas, there was also a consciousness of processes of environmental degradation. Declining soil ferdlity, requiring increasing use of feriliser was listed as a dimension of longterm change in two field sites, and problems in access 1- water was raised elsewhere. l, Coping SlrgWes The experiences of the communities in the 1992 drought provide insights into the nature and effectiveness of community coping mechanisms, household survival strategies and other safety nets, and included the following: [RURAL AREAS) * 'piecework' - i.e. seeking wage labour, generally on the farms of other households (who have conserved food stocks) and paid for in food. * sale of assets: predominantly animals - cattle, goats, chickens etc * food aid: 'aried in effectiveness - highly effective in Simanansa (Southern) and much less effective in Senanga (isolated Western). The official safety net appeared to function well under the circumstances. x  * Wfamlne foods': a variety of tpes of foodstff which are either not normally consumed (such as roots and leaves foraged from the bush), or are not normally seen as critical to the basic diet (such as mangoes) come to have considerable signiflcance during periods of shortage. * 'begging'gilfts to relatives: cited in all field sites as a survival strategy, this response also indicates the finctioning of the community and kinship based systems of support for the vulnerable. [URBAN AREAS] * piecework - for dealing with periodic food shortages and providing a primary resource flow into the household. * reducing food Intake * ooenomning: for example, by supplementng fod supply by cultivating small maize plots; by scavenging for twigs for cooldng fires. - participating in food-for-work schemes. e Idrawing on mutual support networks (although in general, community activities in the urban compounds appeared to be minimal). securing credit - the foremost sources of credit in urban areas appear to be neighbours and kin (primarily to borrow food, such as a tin of maize) and money lenders for cash. * seeking eharity from the church * mpullng chidren out of school * sending children to work * migration: reverse migration was mentioned in some cases as a potential fall back strategy. Some people felt that in the rural areas they would at least have access to land for cultivation, and hence food, whereas in ukban areas such land was unavailable. * petty theft xl  POLICY PIUORITIES PROM THE ZAMIBIA PARTICIPATORY POVERTY ASSESSM ENT lThe overall picture that emerges from discussion of social service and infrastructure provision in the communtes visited is, on the one hand, that of unsatisfactory delivery mechism providing poorer quality products yet operating at increasing cost to the consumer, and, on the other, of fragmentary infrastructure rehabilitation carried out with negligible consultation with community members. Clearly, a first step in improving social services is to upgrade the quality of the planning process, which would necessarily involve consulting and working alongside communities in the design and management of local service delivery mechanisms. The experience of the PPA has shown that rapid appraisal indicators could be developed for identification of priority needs and targeting public expenditures at the local level. Services that are designed around what the poor need, can afford and can maintain have a far better chance of being sustainable. To this end, it will be important to develop flexible structures that link community organisations t. :cal government. A familiar and consistent appeal in the communities was for local government and projects to attend to strengthening the organisational capacity of the urban poor, and not to simply provide goods and services. In terms of the actual welfare of the population, material from the PPA clearly depicts a period of general stress which occurs between November and March in most communities when severe problems of health (malaria), food supply, and heavy expenditure demands (school fees and related costs, purchasing emergency stocks of food, festivals, health fees) combine, in rural areas with the period of maximum labour demand in farming and in urban areas with shortage of money. In rural areas, the stress periods for men and women do not necessarily coincide and the burden of labour is always considerably greater for women. There would appear to be relatively low-cost ways of easing the deluge of expenditures and spreading the burden more evenly throughout the year - for example, by changing the fee cycle for schools to mid-year, by devising a system of progressive tariffs for social services, or by instutionalising phased purchase sche,nes for agricultural inputs, such as fertiser or for health services. Specific sectoral policy priorities that emerge from the PPA are briefly indicated below. 13ducation Epense: the peaking of school related expenditures has been cited as a sigriu;^