╨╧рб▒с>■   -/■   $%&'()*+,                                                                                                                                                                                                                                                                                                                                                                                                                ье┴7 °┐TbjbjUU Цv7|7|╖Є┴      l╕╕╕ЪRl╛╛╛4Єd╫d╫d╫h╠╫ш┘ЇЄИtТш▄fNфXжфжфжфжф&╠ьДPяDjdldldld1Эd╨mg╨=j$v :xьajс ╛ФЁжфжфФЁФЁajР]╛╛жфжфBtР]Р]Р]ФЁВ6╛жф╛жфjdР]ФЁjdР]ЮР].d╛╛.dжф▄▄ аЗ┼&─Єr╤d╫'6.d.d<Xt0Иt.d&y&]j&y.dР]ЄЄ╛╛╛╛┘ ECONOMIC OPPORTUNITIES FOR THE VERY POOR A report of research in Belet Weyne and Mogadishu with recommendations for action A report produced for UNHCR by Richard Davies May 1989 Please note: This file has been converted from Word Perfect. This task is not yet complete. Tables are yet to be re-formated, as of 18/04/04  ACKNOWLEDGEMENTS Authority Approval was given for research in the Mother and Child Health (MCH) Centres by the National Director of MCH's, Dr Cabdi-Kamaal Cali Salaad. Approval for the research in the market places of Belet Weyne was given by Cali Adan Maxammed, Chairman of Administration, Regional Office, Hiran Region. The cooperation and assistance of both the Ministry of Health and the Regional authorities is gratefully appreciated. Funding The research was supported by UNHCR (Somalia) with funds made available under the Income Generation and Area Development programme. Andrew Mayne, Project Officer (Development Planning), provided the initial stimulus and ongoing planning support to the research, as well as ensuring essential funding. His careful analysis, informed comments, and continual interest during all phases of the research have been greatly valued. Author and Research Assistants The author of this report is Richard Davies, who was employed as a Consultant by UNHCR during the period of the research. The planning, implementation and analysis of the research was carried out by Richard Davies, with the assistance of: Ms Naciima Jaamac Yasir, MCH Supervisor, who was provided by the MCH Department as an interviewer and research assistant for the MCH research in Mogadishu and in Belet Weyne. Her concern and conscientious approach added considerably to the value of the information collected. The assistance of Ministry of Health staff in the Yaqshiid and Belet Weyne MCH's with the measurement of children during the interview period should also be recognised. Ms Amina Xassan Barkaan and Mr Cabdiqani Maxammed Yusuf, who were provided thanks to the assistance of Faduma Sharif, National Director of the Family Life Programme, Mogadishu. They carried out all interviews of very small businesses in the markets of Belet Weyne, during a very difficult time of the year. Assistance was also given to them in their work by staff of the local FLP program in Belet Weyne. The assistance given by the Family Life Programme as a whole is appreciated. Support Valuable assistance was also given by the following people: Sharif Sufi Mudhir (UNHCR), Jenny Goodwin, (SCF-UK), Asli Aadan, National MCH office, the staff of CARE Somalia in Mogadishu and Belet Weyne, the staff of the Italian Medical Team, Belet Weyne, and the UNHCR Sub-Office staff in Belet Weyne.  TOC \o "1-4" \h \z   HYPERLINK \l "_Toc70142976" 1. INTRODUCTION  PAGEREF _Toc70142976 \h 23  HYPERLINK \l "_Toc70142977" 2. THE BASIS OF A STRATEGY  PAGEREF _Toc70142977 \h 24  HYPERLINK \l "_Toc70142978" 2.1 THE AIMS OF INCOME GENERATION  PAGEREF _Toc70142978 \h 24  HYPERLINK \l "_Toc70142979" 2.2 LINKING INCOME AND HEALTH  PAGEREF _Toc70142979 \h 25  HYPERLINK \l "_Toc70142980" 2.3 POSSIBLE OBJECTIONS  PAGEREF _Toc70142980 \h 27  HYPERLINK \l "_Toc70142981" 2.4 THE RESEARCH FOCUS  PAGEREF _Toc70142981 \h 29  HYPERLINK \l "_Toc70142982" 3. FAMILIES WITH MALNOURISHED CHILDREN IN BELET WEYNE AND MOGADISHU  PAGEREF _Toc70142982 \h 29  HYPERLINK \l "_Toc70142983" 3.1 METHODOLOGY  PAGEREF _Toc70142983 \h 29  HYPERLINK \l "_Toc70142984" 3.1.1 Areas chosen  PAGEREF _Toc70142984 \h 30  HYPERLINK \l "_Toc70142985" 3.1.2 Measurement procedures  PAGEREF _Toc70142985 \h 30  HYPERLINK \l "_Toc70142986" 3.1.3 The samples  PAGEREF _Toc70142986 \h 31  HYPERLINK \l "_Toc70142987" 3.2 RESULTS  PAGEREF _Toc70142987 \h 32  HYPERLINK \l "_Toc70142988" 3.2.1 MALNUTRITION LEVELS  PAGEREF _Toc70142988 \h 32  HYPERLINK \l "_Toc70142989" 3.2.2 Socio-economic status of households with malnourished children  PAGEREF _Toc70142989 \h 33  HYPERLINK \l "_Toc70142990" 3.2.2.1 Housing conditions  PAGEREF _Toc70142990 \h 34  HYPERLINK \l "_Toc70142991" 3.2.2.2 Household Food Expenditure  PAGEREF _Toc70142991 \h 34  HYPERLINK \l "_Toc70142992" 3.2.2.3 Education  PAGEREF _Toc70142992 \h 35  HYPERLINK \l "_Toc70142993" 3.2.2.4 Length of residence in the city  PAGEREF _Toc70142993 \h 36  HYPERLINK \l "_Toc70142994" 3.2.2.5 Conclusions  PAGEREF _Toc70142994 \h 37  HYPERLINK \l "_Toc70142995" 3.2.3. Income sources of households of malnourished children  PAGEREF _Toc70142995 \h 37  HYPERLINK \l "_Toc70142996" 3.2.3.1 Family structures and income sources  PAGEREF _Toc70142996 \h 38  HYPERLINK \l "_Toc70142997" 3.2.3.2 Income from extended family and others  PAGEREF _Toc70142997 \h 39  HYPERLINK \l "_Toc70142998" 3.2.3.3. Current and previous occupations of mother and fathers  PAGEREF _Toc70142998 \h 42  HYPERLINK \l "_Toc70142999" 3.2.3.4. Occupational mobility  PAGEREF _Toc70142999 \h 44  HYPERLINK \l "_Toc70143000" 3.2.4 Women's attitudes to employment: present and future  PAGEREF _Toc70143000 \h 45  HYPERLINK \l "_Toc70143001" 3.2.4.1. Why women were not earning income at the time (of the interview)  PAGEREF _Toc70143001 \h 46  HYPERLINK \l "_Toc70143002" 3.2.4.2 Child care arrangements  PAGEREF _Toc70143002 \h 52  HYPERLINK \l "_Toc70143003" 3.2.4.3 Desired type of work in the future  PAGEREF _Toc70143003 \h 54  HYPERLINK \l "_Toc70143004" 3.2.4.4 Training and education needs  PAGEREF _Toc70143004 \h 55  HYPERLINK \l "_Toc70143005" 3.2.4.5 Use of a 50,000/- windfall  PAGEREF _Toc70143005 \h 57  HYPERLINK \l "_Toc70143006" 3.2.4.6 Summary  PAGEREF _Toc70143006 \h 61  HYPERLINK \l "_Toc70143007" 3.2.5 CASE-STUDIES OF FAMILIES WITH MALNOURISHED CHILDREN  PAGEREF _Toc70143007 \h 63  HYPERLINK \l "_Toc70143008" 3.2.5.1 Father unemployed and unsupportive  PAGEREF _Toc70143008 \h 63  HYPERLINK \l "_Toc70143009" 3.2.5.2. Father sick with tuberculosis  PAGEREF _Toc70143009 \h 79  HYPERLINK \l "_Toc70143010" 3.2.5.3. Father employed as a soldier  PAGEREF _Toc70143010 \h 80  HYPERLINK \l "_Toc70143011" 4.1 METHODOLOGY  PAGEREF _Toc70143011 \h 84  HYPERLINK \l "_Toc70143012" 4.1.1 The purpose of the research.  PAGEREF _Toc70143012 \h 84  HYPERLINK \l "_Toc70143013" 5.1.2 The sample  PAGEREF _Toc70143013 \h 85  HYPERLINK \l "_Toc70143014" 4.1.3 Data collection  PAGEREF _Toc70143014 \h 86  HYPERLINK \l "_Toc70143015" 4.2. RESULTS  PAGEREF _Toc70143015 \h 87  HYPERLINK \l "_Toc70143016" 4.2.1 Occupations documented  PAGEREF _Toc70143016 \h 87  HYPERLINK \l "_Toc70143017" 4.2.2 ЭФFamily backgroundsЬХ  PAGEREF _Toc70143017 \h 90  HYPERLINK \l "_Toc70143018" 4.2.3. ФAspects of occupationsХ  PAGEREF _Toc70143018 \h 93  HYPERLINK \l "_Toc70143019" 4.2.3.1 Location of businesses  PAGEREF _Toc70143019 \h 93  HYPERLINK \l "_Toc70143020" 4.2.3.2 Equipment  PAGEREF _Toc70143020 \h 96  HYPERLINK \l "_Toc70143021" 4.2.3.3 Supplies  PAGEREF _Toc70143021 \h 103  HYPERLINK \l "_Toc70143022" 4.2.3.4 Daily incomes, capitalisation and profit margins  PAGEREF _Toc70143022 \h 100  HYPERLINK \l "_Toc70143023" 4.2.3.5 Skills and training  PAGEREF _Toc70143023 \h 110  HYPERLINK \l "_Toc70143024" 4.2.3.6. Attitudes towards present and future work  PAGEREF _Toc70143024 \h 111  HYPERLINK \l "_Toc70143025" 4.2.4. Two case studies of small businesses  PAGEREF _Toc70143025 \h 114  HYPERLINK \l "_Toc70143026" 4.2.4.1 Hassan: A Wheelbarrow Trader on Via Mecca, Mogadishu  PAGEREF _Toc70143026 \h 114  HYPERLINK \l "_Toc70143027" 4.2.4.2 Fadumo, a Tea Seller, on the airport road  PAGEREF _Toc70143027 \h 2  5. SUMMARY OF RECOMMENDATIONS FOR ACTION 103 .1 Goals 104 .2 Identifying the very poor 104 .3 Assisting the very poor 105 .1 Specific measures 105 .2 Credit 114 .3 Skills training, technical advice and innovation 118 .4 Monitoring impact 120 .5 Further research 122 6. REFERENCES 123 7. APPENDICES 124 A. Age and sex distribution of samples of malnourished children 125 B. Food prices in Belet Weyne and Mogadishu, late March, 1989 127 C. Comments on experiences with credit programmes in Somalia 128 D. Design modifications for market shelters 138 E. Questionnaires: 1. Families with malnourished children survey 14 2. Low-income occupations survey 145 F. Data-base formats: 1. Families with malnourished children 149 2. Low-income occupations 150 G. Tuberculosis: Demographic profile of 193 patients seen at 151 Belet Weyne TB Out-Patients and In-Patients service, March 1989 1. INTRODUCTION The overall purpose of the research is a practical one: to help direct effective income-generation assistance to the very poor. The focus of the study is on those income-generation activities which can be found in the cities and towns, as distinct from agricultural and pastoral activities, carried out in rural areas. The stimulus for this report has been concern expressed by UNHCR that both past and future developments in the refugee situation suggest a need for improved quality and availability of services to the general population of refugee-affected areas as a whole, particularly to the lowest income groups. This is in contrast to the predominantly camp-based approach to assistance, which has been taken up to now. The contents of this report are based on research that has been carried out in Belet Weyne and Yaqshiid, Mogadishu. Belet Weyne was chosen because it is a large regional town, in the vicinity of a number of refugee camps. Yaqshiid, Mogadishu was chosen for the purposes of pre-testing and comparison. While this report has been written in the context of plans to promote increased involvement of donors and development agencies in the provision of improved services to refugee-affected areas, it is believed that the results will be of value in a wide range of circumstances, not just those affected by refugees. 2. THE BASIS OF A STRATEGY 2.1 THE AIMS OF INCOME GENERATION This report takes a basic position that income generation should be regarded as means to an end, and not an end in itself, as it often is by default. It is particularly important at the planning stages of an income -generation program to give some attention to the purpose behind increasing incomes. Increasing incomes of the very poor may serve any or all of the following purposes: 1. Motivating cooperative activities and mutual support amongst the very poor. 2. Increasing family security through accumulation of savings, either in cash or physical assets. 3. Increasing the choices of goods and services available to the very poor. 4. Increasing consumption of specific goods or services such as food, water, shelter, education or health. 5. Benefiting particular members of the families of the very poor or the family as a whole, for example, women or children, in any of the above ways. The choices which are made will have practical consequences, both in terms of what are the most appropriate means of assistance, and how to monitor the impact of that assistance. A review of credit projects in operation in Somalia in early 1989 shows that relatively little attention was being paid to monitoring the outcomes of credit extension (Davies, 1989). The most common practice was simply to monitor loan repayments. Care Somalia had gone one step further, and regularly tried to monitor the amount of profits generated by the loans. It did not appear that any projects had tried to monitor the impact of the loans on the savings or consumption behaviour of the beneficiaries families. It seems this has happened largely because few agencies have given close or consistent attention to what the goals of income generation should be. 2.2 LINKING INCOME AND HEALTH This report proposes that the basic strategy should be to establish a linkage between income-generation assistance and health services, especially those concerned with children's nutritional status and health. In terms of the various possible income-generation goals mentioned above, the focus should be on children and on increasing poor families spending power on goods and services which are health promoting. The other goals mentioned should not be neglected, but they should be in the service of this basic goal. The reasons behind this view are: 1. Malnutrition and ill health are arguably the most significant output of poverty. High rates of malnutrition amongst children of nomadic families, at the end of a drought, are a direct expression of poverty of income and assets. In less extreme situations, for example in urban areas, malnutrition amongst children is the outcome of a number of influences which are all significantly related to income: food intake, water quantity and quality, sanitation, crowding, access to health services and maternal education level. Malnutrition is only one health outcome of such influences, all of which are affected by household income levels. Others are mortality rates (especially of children), morbidity (sickness) rates and disability. The degree to which income level is a direct influence varies according to the specific health problem. For example household income usually has a more direct link to child rather than infant mortality, and a more direct link to tuberculosis than tetanus. Nevertheless it seems justifiable to regard ill health as the bottom line of poverty - the aspect which we should be most concerned about. Alleviation of family poverty is not being promoted here as the panacea for all health problems. Rather, the belief is that income is a necessary but not sufficient condition. Andrea Cornia's (1984) international review of cross-sectional and time-series studies of factors associated with child welfare, supports this view. 2. Convergence of efforts on health care and income generation will lead to more efficient use of resources. Up to now malnourished children of families affected by drought, of refugee families with insufficient ration cards or other income, and of families who have come to urban MCH centres, have all received World Food Programme (WFP) food rations as their main form of assistance. This assistance is essentially temporary relief, which does not make any change to the underlying income problems of those households. It seems that despite the scale on which such assistance has been given, in no cases has any economic assistance been provided for such families which would enable them to make a sustained increase to their incomes in the longer term. The situation is similar with Tuberculosis In-Patients. They receive some WFP food while they are in hospital, but this is insufficient for the needs of themselves and their dependents. Building in economic support to such a group, both for the period during treatment and afterwards, is essential if the aim is to minimise treatment default rates and to maximise the number of people coming forward to be treated. In both groups (malnourished children and TB patients) income-generation assistance could both address some of the underlying causes of the health problems, and provide an incentive for attendance to the clinics for attention and treatment in the first place. From a review of past experiences with credit extension (Davies, 1989), it seems that many income-generation projects generally reach only a small number of people and often at a high cost per beneficiary. In these circumstances, working with those who are not the poorest of the poor seems difficult to justify, as is the absence of any systematic means of selecting beneficiaries. The benefits of such projects should ideally be targeted to where their impact is most needed. Linking such projects to key health services would be one way of doing so. It should be pointed out that what is not being promoted here is the idea of an "integrated urban development project". Expecting one organisation to deliver services involving more than one sector can easily lead to failure. Convergence simply means the targeting of complementary services to the same target group by organisations who agree on their common objectives. 3. A focus on health dimensions of poverty will make the task of identifying and selecting very poor households more systematic and less contentious than other approaches. Nutrition status can be physically measured. Measurement of a child's weight and height will indicate, when compared to normative data, the degree to which the child is wasted i.e suffering from current malnutrition. Measurement of a child's weight, in relation to their reported age, will indicate the degree to which a child is stunted i.e has suffered from malnutrition over a period of time. Such measurements have already been carried out by Mother and Child Health Centres staff, Refugee Health Unit (RHU) workers, and Nutrition Unit (MoH) and Somali Red Crescent Society staff who visit drought-affected areas. Further training would increase the numbers of staff who are capable of carrying out such work. It should be stated that a single nutrition-status measurement is not a complete panacea to the problem of selecting very poor households. Children from middle-class and wealthy families can easily become severely wasted as a result of bouts of diarrhoea. However, selection error can be reduced by: 1. measuring stunting as well as wasting; 2. measuring nutrition status in all children in the family under 5 years of age; 3. measuring maternal nutrition status. Common sense also suggests that observation of other aspects of families' lives should still be carried out: housing quality, crowding, employment sources etc. Recommendations concerning this stage are made later in this report (Section 5.2). If such measures are adopted as selection criteria, existing institutions such as MCH centres and the RHU health services, can be used to identify very poor households. Additional training would be necessary in the case of many of the MCH's but this would be of benefit to those centres, regardless of the fate of any income-generation activities they might be associated with. It is suggested that such centres would not themselves operate income-generation projects but simply act as agents of referral to income-generation projects. Such an approach would overcome some of the problems with other approaches which agencies have tried. In a number of cases beneficiary identification has been left largely to local authorities, either Camp Commanders or local village committees or elders, on the assumption that they know their community the best. In other cases the implementing agency has sought maximum independence from such people and sought to do all the identification by themselves. Other organisations have found a compromise involving elements of both approaches. In the first instance the risk is that those selected will not necessarily be the poorest, and they will most likely be those who already have a good relationship with local authorities. Income- generation assistance in this situation risks reinforcing or even extending patterns of dependency upon structures which are essentially political. In the second case such agencies, by operating independently of local power structures, risk putting themselves in a position where they can easily be accused of "tribalism", by anyone who is unhappy about the results of their selection. In the longer term there is the problem of the future sustainability of any new administrative structures set up by the agency for the purpose of identifying credit beneficiaries. Nutrition status is an observable and verifiable measure that should be acceptable to all interested parties. Ideally, although health staff might carry out initial measures, the staff of income-generation projects should also be capable of carrying out such measures, both to verify initial measurements and to monitor changes. 2.3 POSSIBLE OBJECTIONS Five objections can be made to the overall approach outlined above: 1. As one senior member of the government has said, "Do we have to wait until people are starving before we can give them assistance?" The answer is quite bluntly no. A proportion of the population, particularly children, are already malnourished, and up to now they have received little but free food provided through MCH's. Consistently targeting what little income-generation assistance is available to those who are not yet "starving" seems an inadequate response while such a situation continues. 2. Other people have pointed out that many very poor families do not use MCH services (Bowker and Keane, 1987). This raises the question of whether it may therefore be inappropriate to use them as points of referral since there is a risk that many of the very poor would not be reached by the associated income generation projects. The answer is that in the short term this may well be the case. However it is expected that a visible linkage, where referral was based on simple and known criteria would, after a period of operation, act as a continuing incentive which would draw in increasing numbers of the poorest families to the MCH centre. This would facilitate the extension of health services to such marginal groups at the same time as it ensured that the income-generation service reached those most in need. Furthermore, other referral links could be established on the same principle, which would increase the numbers of vulnerable children reached through such an approach. These could include referral to income-generation projects from: 1. Banadir Hospital pediatrics section, which is in contact with a large number of poor families, especially those bringing their children there for treatment of severe diarrhoea. 2. TB clinics, especially In-Patients treatment. This report includes some discussion of the family situation of TB patients and the possible income generation responses that could be appropriate (See Appendix G). 3. Using trained and untrained traditional birth attendants (TBA's) many of whom are already in contact with local MCH's. These would require some further basic training, but their major advantage lies in their strong links into the local communities that they serve. 3. Another criticism might be that the approach advocated neglects the critical role of education. However it is recognised that nutrition status, and health in general, is not solely determined by income. Examination of evidence from sources all over the world have shown that education levels, particularly of mothers, has a powerful independent effect of its own (Cornia, 1984). Experience with the operation of income-generation programmes has also shown that training, often very informal, is also great importance to the success of such projects, and that money by itself is often not enough. Education must be built into both health and income generation services. 4. Furthermore, it might be pointed out that malnutrition and poverty are not the same thing. This is very true, but it is worth examining the relationship. Poverty is a broad concept usually defined and measured in terms of a household's overall income or expenditure. Malnutrition can be one of the expressions or consequences of poverty. The specific consequences of poverty, can and do, vary from one family to another, even though their income or expenditure levels may be the same. How people spend (the same amount of) money on housing as opposed to clothing, food, medicine and many other things, will depend on the nature of their income sources, the structure of their family, and their attitudes and understanding. It is not realistic to expect an exact congruence between poverty measured on an aggregate level and any of its specific expressions. If this is accepted, then we should focus on identifying, and working with, those families whose specific expression of poverty is of explicit concern to us, which could be for example those with very poor housing, those with zero school attendance, or those with malnourished children. This would be achieved by using the appropriate specific indicators in each case i.e crowding measures for housing, or anthropometric measurements for nutrition. 5. Finally, it is the case that the aged, and the adult sick and disabled, would not be specifically targeted, and identified in significant numbers, through the use of the referral system that has been proposed so far. There are a number of reasons why the main focus of such a referral system should be on children. Firstly, in terms of indicators, children's health is more indicative of a family's current economic circumstances than the health of the aged, which reflects the impact of a lifetime of experience. Secondly, the argument for giving priority to health indicators relating to children's health is based on the fact that it is this age group (particularly those under 5 years of age) which accounts for the highest proportion of all deaths in the population as a whole. Thirdly, it is far easier to effect the causes of morbidity and mortality in this age group than it is amongst the aged. Nevertheless, where services for the aged and the disabled (in particular) are in operation, it would be appropriate to investigate the value of establishing, through referral, a supporting programme of income generation assistance. 2.4 THE RESEARCH FOCUS This report is based on two main pieces of research. One consists of two surveys of women with malnourished children brought to MCH centres in Yaqshiid, Mogadishu and Belet Weyne. The other consists of a survey of men and women involved in very small businesses in the markets of Belet Weyne. The results of these surveys have been supplemented by case studies, most of which have been carried out in Mogadishu, as well as information from existing income-generation services, especially those operated by CARE Somalia. An additional smaller piece of research was done with TB patients, also within Belet Weyne. This research has been carried out with two questions in mind: 1. How can projects identify and differentiate those groups to whom income generation assistance should be targeted? 2. What forms of assistance are appropriate to those groups that have been differentiated? In addition to the strong focus on the health dimension of poverty, this report has also given a strong emphasis to a family perspective. Information on occupations has been gathered with the assumption in mind that the significance of income available from any occupation can only be understood within an understanding of the family context. John Iliffe (1987,p.8), author of "The African Poor: a history" has commented: "The intimate connection between poverty and family structure has been neglected by historians of Europe and may be Africa's chief contribution to the comparative history of the poor" While this research was not intended to be an anthropological study, the attempt has been made to document aspects of the family structures of those interviewed. It is hoped that this will move discussion about vulnerable or priority groups further on from broad statements, for example, about female headed households, towards an fuller appreciation of the diversity of family structures and their different implications both for the survival of the families themselves and for the selection of appropriate forms of economic assistance. 3. FAMILIES WITH MALNOURISHED CHILDREN IN BELET WEYNE AND MOGADISHU Between January and April 1989, information was gathered from 90 mothers with malnourished children who had been brought to two Mother and Child Health Centres (MCH's) in Belet Weyne and in Yaqshiid, Mogadishu. The methods used and the results of these surveys are described below. 3.1 METHODOLOGY 3.1.1 Areas chosen The Belet Weyne MCH was chosen because it was the main point of health service for women with children under 5 years of age in Belet Weyne, the area this study has focused on. The Yaqshiid MCH was chosen out of 15 MCH's in Mogadishu, for two reasons. Firstly, according to the National MCH Director, that MCH was functioning well and the staff were capable of carrying out the height and weight measurements that were required. Secondly, the area was adjacent to Heegan obbosibo (squatter area), a poor area of Mogadishu where Oxfam had begun income-generation work with poor families in 1988. It was hoped some of the information obtained from the Yaqshiid MCH interviews could be utilised in a practical way by the Oxfam staff. It was also thought that some of the poorest women interviewed could be referred to Oxfam for inclusion in their project activities. The Mogadishu interviews were carried out both for the purpose of pre-testing the questionnaire and to allow some comparison with the findings of the Belet Weyne survey. 3.1.2 Measurement procedures In each MCH an attempt was made to measure the height and weight of all children attending. This was done by two MCH staff who had been given one morning's revision training in how to carry out such measurements. A table-top beam balance scale was used for the youngest children and a free-standing floor scale was used for the older children. The heights of most of the children were measured on a measuring board laid on a table top, but older children were measured while standing on the floor scale. Height measurements were made to the nearest half-centimetre. Calculations of percentage of normal weight-for-height were made by reference to the NCHS/CDC/WHO normalised reference. The age-of-the child was also recorded, where possible, to the nearest month. Weight for age was calculated to the nearest 1% using weight-for-age information published by WHO (1983). Accuracy of weight measurement was generally good, with the base setting of the scale being checked most mornings!. Accuracy with length measurements was less adequate. The most common error was to leave the legs still bent, or the head not straight, thus underestimating the child's length. Age was more of a problem, since after two years of age most mothers did not give their child's age in months to any very precise degree. Interviews of mothers of the children in Belet Weyne and Mogadishu were carried out by Naciima Jamac Yasir, an MCH Supervisor. In Yaqshiid, interviews were carried out in an adjacent MCH room where privacy was available. In Belet Weyne less privacy was available, and most interviews had to be carried out in the open courtyard outside the MCH centre. At the beginning of the survey in Yaqshiid, a two-day period was spent doing trial interviews of staff and women at the MCH. As a whole, the quality of the interviews was good, although much relevant detail was not written down in the first instance. All interviews were checked at the end of each morning, and the interviewer herself was subsequently interviewed in order to ensure that all known information was recorded. A copy of the interview schedule is included in Appendix E.1. It is hoped that this questionnaire, or one based on it, will be used in future by any agency interested in the establishment of referral links between MCH centres and income-generation projects. 3.1.3 The samples In YAQSHIID, measurements of children and interviews of their mothers were carried out between January 22nd to February 5th. A total of 340 children were brought to the centre, of which 286 (84%) were measured. Average daily attendance during this period was 31 children, compared to an average of 26.4 per day during the whole of January and 32 per day in December 1988. The remaining 13% of children were not measured because of a combination of reasons. On some days the MCH staff were too busy to ensure that every child was measured. There was also an initial misunderstanding by staff that children who were obviously well nourished did not need to be measured. Despite the fact that not all children were measured, from observation it seemed that almost every malnourished child who came to the MCH was in fact measured. Of the 45 women with malnourished children, a total of 41 were interviewed. No women refused to be interviewed but four were missed out because they were not referred on to the interviewer by the MCH staff, or they were simply lost in the crowd. In BELET WEYNE, measurements of children, as well as interviews of their mothers, were carried out from February 16th to March 1st, and from March 18th to 23rd. A total of 330 children were brought to the centre, of which 283 (86%) were measured. Average daily attendance during this period was 17 children, compared to 15 during the first three months of 1989 as a whole. As with the YAQSHIID MCH, the main reason why the remaining 14% were not measured was pressure of numbers on the busiest days (Saturdays). Of the 63 women with malnourished children, a total of 49 were interviewed. One women refused to be interviewed and another interview with a father of a child was cancelled because of very limited cooperation. A further two were not interviewed because their children were referred immediately for rehydration. The remaining 10 were lost in the confusion! The age and sex distribution of all the children who were measured, and those whose mothers were interviewed is shown in Appendix A. 3.2 RESULTS 3.2.1 MALNUTRITION LEVELS The incidence of malnutrition in children attending BELET WEYNE and YAQSHIID MCH's is shown below. %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% % Children measured % Normal weight-for-height: % % %%%%%%%%%%%%%%%%% (a.)81-85% (b.)80% or less % % % %BELET WEYNE 285 33 37 % % = 11.6% = 12.9% % % % %YAQSHIID 340 33 19 % % = 9.7% = 5.6% % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% In the case of the YAQSHIID MCH the above figure is possibly a slight overestimate of the normal incidence. On four days, TBA's attached to the MCH were asked to find mothers with malnourished children and to ask them to come to the MCH. However, despite efforts that were made, accurate records were not kept of all of those children who came to the MCH on this basis. Our very approximate estimate is that perhaps 40 children came to the MCH this way. On the basis of observation alone, it seemed that many of the children who were referred by TBA's were little different to those already coming. At the most, one in five of the children referred may have been 85%, or less, of normal weight-for- height. Bearing in mind that the incidence of malnourished children (80% of normal WFH or less) in Mogadishu as a whole is probably between 2.5% and 6% of children under 5 years (UBS, 1988) the number of malnourished children being brought to the YAQSHIID MCH is hardly impressive. For a group such as this, which is particularly vulnerable an MCH should ideally be reaching a far higher proportion than exist in the community as a whole. Projections of the total number of malnourished children coming to the MCH in the course of a year can be made. However they need to take into account the fact that some the of children brought to the MCH's are making a return visit (for immunisation, WFP food rations and other reasons). Examination of the YAQSHIID MCH register during the survey period showed that 41.4% of the children being presented at the MCH for the first time. An ad hoc examination of one days sample of malnourished children showed a similar proportion. If this is normally the case, then it could be expected that with daily attendance rate of 30 children, a total of approximately 212 malnourished children (80% of normal WFH or less) would come to the attention of the MCH in the course of a year. In the case of BELET WEYNE MCH the proportion of children coming to the MCH for the first time was much higher - 77% of all the children measured. This is not surprising since that MCH appeared to be not as well managed as the YAQSHIID MCH. Given an average daily attendance rate of 17, this could suggest a total of approximately 617 malnourished children (of 80% of normal WFH or less) would come to the attention of the MCH within one year. It could be argued that some of these cases of malnutrition are episodic, a result of severe diarrhoea or other illness, and are not sufficient grounds for extending income-generation assistance to those families. However examination of the ages of the children in these groups showed that the average weigh-for-age was 70% of normal in the YAQSHIID group, and 73.8% or normal in the BELET WEYNE group. This suggests that both groups of children have in fact been malnourished for an extended period beforehand. Measurement of nutrition status of siblings and mothers would also clarify the status of the family, as already suggested. If additional numbers of malnourished children are needed to justify the establishment of an income-generation link with a particular MCH, the simplest method would be to use the attached Traditional Birth Attendants. While their success in Yaqshiid was only marginally better than what was occurring with the existing flow of women into the MCH, with simple training and explanation of purpose it is likely that these women could become quite effective in identifying malnourished children in the community. Their reward would be the opportunity to mediate significant opportunities to people within their community. It should be noted that in the case of both the Yaqshiid and Belet Weyne surveys, the names of the mothers have been recorded along with those of the children who were measured. In addition the names of the xaafa (village) and, in Belet Weyne, the waax (section) in which they were living have also been recorded. It should be possible to make contact with them again in the future, either to identify what changes have happened in the children's nutritional status, or to include those families into any income generation projects which have since initiated in those areas. 3.2.2 Socio-economic status of households with malnourished children Information on housing conditions, household food expenditure, and educational status of the households from YAQSHIID and BELET WEYNE, have been summarised below to give an indication of their relative socio-economic status. 3.2.2.1 Housing conditions Housing, particularly crowding, is an important aspect of quality of life. In urban areas it is also normally correlated closely with household income levels. In the course of the interview, women were asked how many rooms their household used and earlier on, the size of their household at that time. The aggregated results are shown below: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Housing Average Number Households living in % % People per Room one room % % % %Mogadishu 1986: % % % % - Average area 2.8 30% % % % % - Yaqshiid(squatters) 3.5 49% % % % %YAQSHIID MCH 4.6 58% % % % %BELET WEYNE MCH 5.2 65% % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 3.2.2.2 Household Food Expenditure As an approximate means of assessing household expenditure level' the interviewer asked each woman how much money her household spent each day on daily food expenses (known in Somali as masruuf or biil). The aggregate results for the two groups of women were as follows: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Daily food expenditure Median Equivalent in maize, Kg per head % % per head (Food Early Warning System prices) % %Mogadishu 1986: % % % % - Average (Waaberi) 50/- 1.96 kg (November) % % % % - Yaqshiid (squatters) 37.5/- 1.27 kg (mid July) % % % %YAQSHIID MCH (1989) 75/- 0.83 kg (late January) % % % %BELET WEYNE MCH (1989) 60/- 0.45 - 0.63 kg (late Mar, late Feb)% %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Equivalents in maize value, based on FEWS prices in each area at the time of the respective surveys, have been given in an attempt to make the expenditure levels comparable despite the impact of inflation. Maize has been used instead of rice because it is a food item which is used very widely by the poorest households in the south of Somalia. It should be noted that purchasing power, as indicated above, may vary through the year because of seasonal variations in maize prices. Some caution in interpreting the figures is also required because, in some cases, people may simply be understating their levels of expenditure, and some food items consumed may not have been purchased (e.g milk and eggs, in both Yaqshiid and Belet Weyne). To gain an impression of total expenditure it can be noted that food as a percentage of total expenditure has been found to be on average 56.1% in Mogadishu in 1985 and 80% in Kismaayo in 1987. The percentage normally diminishes with increases in income levels. 3.2.2.3 Education The levels of female literacy and primary school attendance in YAQSHIID and BELET WEYNE households were as follows: (see table next page) %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Education Adult female % of children of primary school% % literacy age attending school % %Mogadishu 1986: % % % % - Average area 47% 54% % % % % - Yaqshiid (squatters) 26% 33% % % % %YAQSHIID MCH 17% 23% % % % %BELET WEYNE MCH 21% 19% % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% As with the above quoted Mogadishu surveys, and other major demographic surveys carried out in Somalia in recent years, women were asked if they could read and write but were not asked to attempt any simple test. Thus the above percentages are generous estimates. It is possible that many who said they were literate could not do more than read and write their own names. 3.2.2.4 Length of residence in the city In the YAQSHIID sample, 22% of the household heads had been born in Mogadishu, and of those who had migrated to the city the average length of residence was 17 years. The vast majority of families were not new migrants to the city. Only 2 household heads had come to the city within the last 5 years (5%). Compared to those living in the adjacent squatter area, more of the household heads in the MCH sample had been born in Mogadishu, and of those who had migrated the average length of residence was slightly longer (Davies, 1988, p.263). In the BELET WEYNE sample, 47% of the household heads had been born in Belet Weyne, and of those who had migrated to the town the average length of residence was 11 years. Of the migrants, 22% had come within the last five years, and of these most within the last year. 3.2.2.5 Conclusions 1. At least on the above indicators it does seem that the sample of women interviewed in YAQSHIID MCH are more disadvantaged than those that can be found in a random sample of an average income area in Mogadishu, or of a squatter area in the immediate vicinity of the YAQSHIID MCH. This supports the view that screening for families with poor nutrition status through an MCH in Mogadishu is a relatively effective means of targeting programs of assistance to some of the most disadvantaged in that community. 2. Since there is no normative data on Belet Weyne, it is not possible to make the same comparison there. However, compared to the YAQSHIID women, the BELET WEYNE women are more disadvantaged in terms of housing and food expenditure and education (children's school attendance). 3. In all three above tables the figures quoted are medians. Thus it should be remembered that 50% of the families will be worse off than these figures suggest, and 50% will be better off. There is no doubt that in both MCH groups there were some families that were not poor. The same is the case, with the sample taken from Yaqshiid squatter area in 1986. In both cases services targeted to the poor found through such approaches would still have to do a further filtering out of some who are evidently not so poor. The most appropriate second stage would be a visit to the homes of those identified by initial height and weight measurement of one of their children at an MCH centre. At that stage, height, weight and age information could be collected on all the children in the family, as well as general information about the household composition and income sources. 3.2.3. Income sources of households of malnourished children In the section that follows, the income sources of the households have been described from four perspectives. With each an attempt has been made to identify the implications for the delivery of income-generation assistance. 1. Family structures and income sources; 2. Income from extended family and others; 3. Current occupations of men and women in the family; 4. Occupational mobility 3.2.3.1 Family structures and income sources YAQSHIID MCH SAMPLE: %%%%%MOTHER%%%%%% % % present absent %%%%4%%%%%% % % % % contributing yes no no income ? % % % T%P%P%P%P%9ЁP%P%P%╨P%P%P%P%9ЁP%P%P%f%P%P%P%P%P%P%9ЁP%P%P%P%P%P%W% Q% 4 %22 Q% 0 Q% %%yes%%%%%%%┬ % Q% Q% 26 64% % Q% 10%% 54%Q% 0%Q% %%present ├%%%%%%%%<%%%%%%%%%╬%%%%%%%%%%%%%┬ % % Q% 5 % 2 Q% 0 Q% % %%no%%%%%%%%┬ % Q% Q% 7 17% % Q% 12%% 5%Q% 0%Q% FATHER `%P%P%P%P%P%P%P%P%╧P%P%P%P%P%P%P%P%l%P%P%P%P%P%P%P%P%P%P%P%P%P%c% % Q% 0 % 3 Q% 0 Q% % %%yes%%%%%%%┬ % Q% Q% 3 7% % % Q% 0%% 7%Q% 0%Q% %%absent ├%%%%%%%%<%%%%%%%%%╬%%%%%%%%%%%%%┬ % Q% 3 % 0 Q% 2 Q% %%no%%%%%%%%┬ % Q% Q% 5 12% Q% 7%% 0%Q% 5%Q% Z%P%P%P%P%P%P%P%P%9ЁP%P%P%P%P%P%P%P%i%P%P%P%P%P%P%P%P%P%P%P%P%P%]% 12 27 2 41 29% 66% 5% 100% BELET WEYNE MCH SAMPLE: %%%%%%MOTHER%%%%% % % present absent %%%%4%%%%%% % contributing % % % income ? yes no no % % % T%P%P%P%P%9ЁP%P%P%╨P%P%P%P%9ЁP%P%P%f%P%P%P%P%P%P%9ЁP%P%P%P%P%P%W% Q%10 %23 Q% 3 Q% %%yes%%%%%%%┬ % Q% Q% 36 73% % Q% 20%% 47%Q% 6%Q% %%present ├%%%%%%%%<%%%%%%%%%╬%%%%%%%%%%%%%┬ % % Q% 2 % 1 Q% 0 Q% % %%no%%%%%%%%┬ % Q% Q% 3 6% % Q% 4%% 2%Q% 0%Q% FATHER `%P%P%P%P%P%P%P%P%╧P%P%P%P%P%P%P%P%l%P%P%P%P%P%P%P%P%P%P%P%P%P%c% % Q% 2 % 3 Q% 0 Q% % %%yes%%%%%%%┬ % Q% Q% 5 10% % % Q% 4%% 6%Q% 0%Q% %%absent ├%%%%%%%%<%%%%%%%%%╬%%%%%%%%%%%%%┬ % Q% 1 % 2 Q% 2 Q% %%no%%%%%%%%┬ % Q% Q% 5 10% Q% 2%% 4%Q% 4%Q% Z%P%P%P%P%P%P%P%P%9ЁP%P%P%P%P%P%P%P%i%P%P%P%P%P%P%P%P%P%P%P%P%P%]% 15 29 5 49 31% 59% 10% 100% The two tables above suggest that plans for assistance with income generation will need to take into account the range of family structures and income sources that exist in the poorest households (even when support from relatives is temporarily ignored). While female-headed households have received some attention there are other structures which are also disadvantageous, especially for children. The main groups which were found are described below, starting with those who appear to be the most disadvantaged, and who may also be the most difficult to assist. A. Children solely in the care of grandmothers. Four of these were found in the two samples. These families seemed particularly destitute. Examples: - Abdi (5yrs) and Xalimo (1 year) 78% and 79% of normal weight-for-height respectively, were in the care of their maternal grandmother aged 50yrs. She used to sell charcoal but does no longer does so because she is sick. This family of three is supported in kind by relatives within Mogadishu who bring meals to them each day, to their one-room cariish (mud-walled house) in Yaqshiid. The mother of the children was divorced and is now living "in the bush". - Ali (14mths) 77% of normal weight-for-height, was in the care of his maternal grandmother(46yrs), along with his brother (not measured). The family of three were dependent on support from relatives in the town and their neighbors. His mother died and the whereabouts of the father is unknown. - Faysal (3yrs) 71% of normal weight-for-height, was in the care of his maternal grandmother, along with his 10yr old sister (not measured). The grandmother (aged 60yrs) works as a farmer near Belet Weyne. Her husband had died in the previous month. The mother and father had divorced, leaving the children with the grandmother. - Leyla (1yr) 83% of normal weight-for-height was in the care of her mother's sister (aged 19yrs), her paternal grandmother (65yrs) and her uncle. The family of four was supported by the uncle who sells charcoal. The mother is divorced and living elsewhere. The child showed signs of neglect. Though the absence of parents and other persons of usual employable age appears at first sight to render employment support to the family difficult, there are in fact a few ways in which income-generation assistance might be channeled to such households: 1. To Ali's grandmother, who is still young enough to work. No other immediate options are available. 2. In the case of Abdi, Faysal and Leyla it is not clear why the mother has not continued caring for the child. In some cases it could be appropriate to offer income-generation assistance to the mother, conditional upon support of the child. In other cases where the mother has remarried, the new husband may not have wanted her previous children and this option would not be feasible. 3. In other more difficult cases it may be appropriate to offer income-generation assistance to friends or neighbours of the grandmother, conditional upon their support of the grandmother and child. Those already giving some modest support could be a suitable choice. With the second and third option it is obvious that care and sensitivity would be needed in establishing their feasibility. B. Children whose parents have divorced, and who are now in the care of their fathers and paternal grandmothers. Three of these were found in the two groups. Examples: - Maxammed (6mths) 78% of normal weight-for-height, is being cared for by his paternal grandmother. They, along with two of the fathers younger brothers are supported by the fathers income from collecting and selling firewood. The mother was divorced and now lives elsewhere. - Nimco (8mths) 84% of weight-for-height, is a twin and was cared for by his paternal grandmother (aged 60) and his father who is a Koran teacher. The mother is divorced and living elsewhere. - Maxammed (12mths), 50% of normal weight-for-weight (immediately hospitalised), was being cared for by his paternal grandmother (70yrs) and 10yr old sister. They, along with four other family members were supported by the fathers income from a shop and an uncles income as teacher. The mother is divorced and now living elsewhere. In these cases income-generation assistance could be directed to the fathers, but consideration should also be given to how health education could be passed on at the same time. The aim would be to encourage the men, perhaps against tradition, not to leave all responsibility for the child's care in the hands of the grandparent, who is least likely to be open to new ways of dealing with children's health problems. C. Children living in female-headed households. These were by no means a single group with common characteristics. There were two main types, with divisions amongst each of these: 1. Those (8) where the father was living elsewhere but sending support back to the family. These fell into two groups: - those whose husbands were earning income elsewhere: two drivers, two soldiers and one husband working in Jeddah. In the case of the two soldiers, from Belet Weyne, the mothers were each earning income themselves. - those divorced women whose ex-husbands were sending them support (3 in Belet Weyne). Two of these were living by themselves, and the third, a younger women, was living with her parents, brothers and sisters. None of the women were earning income. In the case of the first group Asli Aadan, of the MCH Department of the Ministry of Health, has pointed out that women whose husbands are elsewhere and sending money back to the family often take the opportunity to use that money to make some savings for themselves, even though that may mean cutting back on food consumed by themselves and their children. In other cases, such as the families of soldiers, that opportunity is not likely to exist and women need to find their own source of income. In the case of the second group there is the risk that income-generation assistance could end up displacing the support provided by the ex-husbands (See Section 3.2.5.1 for a case study of a situation of this type). 2. Those (6) where the absent father were not sending any support. These were in two sub-groups: - those (4) where the mother herself was earning income. Three were widowed women with children, living by themselves n Yaqshiid, earning income by making mats and muufo. One was a younger divorced women with children who was living with her parents, brothers and sisters in Belet Weyne. - those (2) where the mother was not earning any income. In both these Belet Weyne families the mothers were young, divorced and living with their parents, brothers and sisters. It should be noted that the proportion of female headed families found in these two groups was no larger than that found in average-income areas of Mogadishu, in research carried out in 1986 (Davies, 1988). This supports the view, expressed in the same research report, that in urban areas of Somalia all female-headed households cannot automatically be considered to be a disadvantaged group. D. Children living with both parents. Not surprisingly, these were the largest group (33 in Yaqshiid, 38 in Belet Weyne). These consisted of 4 sub-groups: 1. Those where neither parent was earning any income (3). In one case the father has TB and the mother has two very young children to care for. In another the husband has neglected the family, giving only occasional support, obtained from "shaxaad" (handouts). In the third family, in Belet Weyne, the husband is a high-school graduate, waiting for work, pending which the family is supported by relatives. 2. Those where the mother was the sole income earner (7), although the father was present. Five of the men were aged (60+), one young man was sick and the other unemployed. 3. Those where both mother and father were earning income (4 in Yaqshiid and 11 in Belet Weyne). 4. Those where the father was the only parent earning income. This was the largest group of all and accounted for approximately half of the families interviewed (54% in Yaqshiid, 47% in Belet Weyne). The type of income-generation assistance which can be extended to this group will partly depend on the attitudes of the mothers themselves to earning income for the family. Data collected on the attitudes of women from this type of family are described and discussed in Section 4 of this report. There are at least three other aspects of family structure which have consequences for family incomes and their distribution within the family. The first is the age relationship between the mothers and fathers. In Belet Weyne the average age difference between mothers and fathers was 11 years, but in 6 families the fathers were 20 or more years older than the mothers. The potential disadvantage this poses is that when the father grows old he will become dependent on the family for some years before he dies. This is not necessarily a problem if the couple started to have children many years ago and some of the children have since grown to working age. However if it is a second or later marriage the children may still be quite young during the period when the father is aged and out of work. At this stage a major burden falls on the mother, and her extended family, if support is available from them. Five of the families interviewed were in this situation. It could be argued that these households are in a worse position than conventionally-defined female-headed households. The second is the impact of polygamy. A cursory analysis of the Belet Weyne families showed that 11 of the 39 men currently married had second wives. The extent to which they were supporting these families was not systematically investigated, but in many of these cases it did seem that the other families were receiving some support from the husband. The third is the impact of divorce and re-marriage, through its effects on the use of income within the family, especially on behalf of the children. In 14 of the 39 Belet Weyne families having both mothers and fathers present, either the mother or father had been divorced in the past. In many such cases the children from the previous marriage may be integrated into the current family. If these children are the husbands, there is risk that the new wife will give her own children of this marriage greater priority with feeding and care in general. Alternatively the husband's children may be passed on to his sisters or mother to be cared for. Street children - those who actually live as well as work on the streets - are one manifestation of the break-up and re-mixing of families. Research in Mogadishu in 1986 showed that most of the street children had left such families. It would be useful if more attention could be given to identifying the effects of family breakup and reconstruction on the welfare of both women and children in Somalia. 3.2.3.2 Income from extended family and others Support from extended families and others (excluding husbands working elsewhere) was received by 27% of families in Yaqshiid and 32% of families in Belet Weyne (within the previous month). This compares to 41% of families in the Yaqshiid squatter area (1986) and 70% of families in Waaberi (1986), a middle-income area of Mogadishu. Amongst those families where there was no husband providing income, at least half were receiving support from others: 50% (6 of 12) in Yaqshiid and 62.5% (5 of 8) in Belet Weyne. The most common sources of support for all the families interviewed were as follows: (See table next page) Sources of support received by families with malnourished children %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%% %Sources of support: (in relationship to% Yaqshiid % Belet Weyne % %mother and excluding supoort from % % % %husbands living elsewhere) % # reported*% # reported* % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%$% %1. Immediate family % % father 5 % % brother 1 3 % % mother 1 2 % % sister 1 % % % % Sub-total 2 11 % %2. Other relatives % % undefined 3 4 % % aunt 2 1 % % % % Sub-total 5 5 % %3. Others % % neighbours 2 1 % % friends 1 % % persons o/seas 1 % % refugee rations 3 % % % % Sub-total 4 4 % % % % Total receiving support 11 18 % % % % * some families received support from more than one source % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% In both "developed" and "developing" countries, social isolation is an important dimension of poverty. A longer-term goal of income-generation assistance should be to decrease the isolation of the poorest and increase their access to social support networks. In doing so a dual perspective is needed. One the one hand beneficiaries' existing social support networks must not be accidentally undermined, as for example in the case of assistance received from ex-husbands, or for small businesses, the availability of credit/goods on consignment from other businesses. On the other hand, new forms of security through cooperation with others should be encouraged, even though many beneficiaries will be operating businesses on an individual basis. One way of preserving what is already of value while developing new forms of support is to adapt the function of rotating credit groups known as hagbad, shalongo or ayuto. Such groups are a means of making savings when there is intense pressure to spend whatever money is available, and a means of obtaining sufficient capital to make major purchases. Typically, an informal and self-selected grouping of women each contribute an equal amount to a common fund on a periodic basis (for example, each day or week). In turn, the money in this accumulated fund is periodically distributed in a lump sum, to each member in turn. Despite the fact that the groups can only function on the basis of trust and fair play many such groups continue to function for many years (see Davies, 1988 p.168). As Care Somalia has shown it is only a short step between this type of structure and the extension of loans to people through small group structures. Group-based credit extension has been recognised by many, including the World Bank (1988,p.6), as an appropriate means of delivering credit to large numbers of poor households who have minimal collateral. The model people are most familiar with is that used by the Grameen Bank in Bangladesh or that use by the Working Women's Forum in Madras. In both cases there is no insistence that group members carry out a single economic activity together. The most common functions of members of such groups are to assess each others' loan requests and to act as guarantors for each others' loans on a collective basis. Apart from their economic value an equally important reason for using such an approach is the opportunity such groups present for pursuing social-support and development goals at the same time as credit is being extended. For example, within the overall strategy advocated in this report such groups would provide an alternative means of communicating key health messages to a particulary vulnerable group. They would offer the opportunity for mutual learning amongst the members about their common health problems, as well as an opportunity for health service staff themselves to learn about the women's own perspective. At present the main means of communicating health information is on an individual basis, as each woman waits in a crowd to obtain treatment for her child's immediate health problems at a local MCH. With credit groups a critical issue is the process of group formation. How they are developed can influence their effectiveness from both a financial and social point of view. There are two basic requirements that must be met but are not always easy to combine: 1. The members must belong to the target group that assisting organisation wants to work with. 2. The members of the groups must be made up of people who are happy to work with each other. One approach used by some organisations like Oxfam has been to respond to, or try to find, such groups which are already in existence. However this can be a slow process, partly because many of those who are of concern are not even at the most basic stages of self-organisation. A more active approach can be taken by intervening, but where possible building on existing social relationships. In a previous paper a number of possible approaches to group formation have been examined (See Appendix C). One approach to the development of cohesive groups of people belonging to the target group could be as follows: 1. Bring women who are eligible group members to meet together e.g. mothers of malnourished children. 2. Ensure that the women come from a relatively small geographic area, e.g a xaafa (village), so that some of them are likely to already know each other. 3. Allow the women present to sort themselves into groups of their own choosing, according to whom they know best, with no minimum size to the groups. 4. Because a number of women will not know each other, and some of the groups may not yet be large enough (e.g 5 or less), the women should then be allowed to include into their groups a certain number of other women they know from outside the meeting, on the condition that they do fit the membership criteria i.e they have children who are malnourished. The type of other women included on the basis of friendship may vary. In Yaqshiid, 11 of the 41 women said they knew other women with malnourished children. In Belet Weyne 3 of the 49 knew of others. With the incentive of being able to form a group which might capture some benefits for them, a larger number might be found. However, interpretations of who is malnourished might be liberal, as was the case with TBA's used in the Yaqshiid research, unless some type of orientation is given beforehand. For the purpose of group formation some tolerance over the definition of malnutrition would be practical in any case. 3.2.3.3. Current and previous occupations of mother and fathers (See table next page) .1 Fathers in YAQSHIID MCH SAMPLE (37) %%%%%,%%%%%%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%%%%% %Age % Current Employment % Past Employment % Reason for change % %%%%%4%%%%%%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%%%%$% %70 unemployed watchman age+sickness % %60 unemployed watchman work finished % %75 unemployed watchman sickness % %25 unemployed watchman sickness % %30 unemployed porter TB % %45 unemployed quarry labourer accused of theft % %31 unemployed guulwade (Militia) may still be % % % %45 soldier none * % %35 soldier same % %30 soldier same % %50 soldier same % %32 soldier farmer not enough income % % % %32 garage worker none % %36 bus- money collector * * % %36 mini-bus driver none % %35 truck driver to/from Balcad waiter in restaurant money too small % %50 truck owner/driver same % % % %40 Local Govt * * % %30 Local Govt none * % %40 Govt-central bank soldier transferred % % % %30 house builder muufo dough maker * % %50 house builder same % %30 laborer in the Port same % %40 labourer in the market operated sesame press * % %25 labourer in quarry same % %40 labourer in quarry maize grinder ADC cut profits % %40 measures charcoal at Co-op pastoralist came to Mogadishu % % % %40 watchman govt office worker Som/Ethiopia war % % % %55 plastic bag seller * * % %35 handcart trader-miscell clothes seller fire burnt out shop % %40 dry rations shop pastoralist came to Mogadishu % %40 sell rice and pasta in market soldier did not like it! % % % %50 haircutter same % %40 koran teacher same % % % %40 farmer Koran teacher * % %45 fisherman same % % % %24 overseas carpenter not earning enough % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% * = not known .2 Fathers in BELET WEYNE MCH SAMPLE (44) %%%%%,%%%%%%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%%%% %Age % Current % Past % Reason for change % %%%%%4%%%%%%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%%%$% %65 unemployed Ministry of Education Retired (age) % %20 unemployed none Ex-student % %60 unemployed porter mental illness % %60 unemployed collected money on trucks Som/Ethiopia war % %30 unemployed koranic teacher Fighting in north % % % %48 soldier same % %28 soldier same % %30 soldier same % %45 soldier same % %* soldier same % %40 soldier same % %40 soldier lived in the bush wanted to be soldier% %36 soldier farmer wanted to be soldier% %35 soldier pastoralist wanted to leave bush% %45 soldier pastoralist wanted to be soldier% %35 soldier carpenter conscripted % %25 soldier carried wood to sell * % %30 soldier built houses * % %62 animal broker same % %40 animal broker animal trader lack of money % %50 animal broker house builder work finished % %40 animals - in BW soldier TB % %50 animals-buys/sells shop selling food family ate capital % %40 animals-buys/sells * * % %35 pastoralist tea shop he likes the bush % % % %50 govt-watchman cook Som/Ethiopia war % %35 local govt farmer sickness % %35 local govt pastoralist came to Belet Weyne % %34 govt-refugee office teacher promotion % %* govt-ADC Mogadishu * * % % % %30 donkey cart operator none * % %38 donkey cart operator soldier mental illness % %30 money collector(truck) rode truck as helper * % %60 money collector(truck) farmer no rain % %50 driver of Pick-up * * % % % %35 biibito (cold drinks) * * % %35 clothes seller * * % %35 clothes seller * * % %30 shop selling food items soldier too little money % %20 Koranic Teacher * * % %50 building houses * * % %34 laboratory worker translator work finished % %* overseas (nurse) * * = not known % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% 3.2.3.3.1 Observations: The level of unemployment amongst the fathers of the children (those working elsewhere included) was relatively low: 16% in Yaqshiid and 11% in Belet Weyne. This is consistent with survey findings in Mogadishu, which have found the lowest incidence of unemployment in the poorest areas. Of the 12 unemployed, six were aged 60 years or more. Of the others one was sick, one was recently displaced by the fighting in the north and looking for work, one was an ex-student looking for work. Three others were living on "shaxaad" (asking friends and acquaintances for money on a day-to-day basis). Like unemployment benefits in the U.K or other countries, shaxaad can be a useful safety net or an alternative, and often perceived as undesirable, means of making a living over a long period of time. When targeting income generation assistance it would be inappropriate to view unemployment as an indicator of disadvantage. In most cases, particularly men with families to support, it is more appropriate in a country with no unemployment benefits, to interpret unemployment (other than short term) as something that a person has been able to afford. In the section below, comments have been made on each of the main types of income sources that were reported. A. The military: When compared to the incidence of men employed in the armed forces in Mogadishu (approximately 5%) the incidence in both groups appears very high (15%-27%). In the case of Belet Weyne (27%) this may be partly explained by the fact that it is in a sense a garrison town. However it is useful to recollect some general features of armies. In many countries, armies have offered employment to many men with minimal education. This was especially the case in the early years after independence in Somalia, when a major part of the government work force were soldiers. In Somalia the bulk of those who have joined the army voluntarily have come from farming and pastoral backgrounds. The army has been a step upwards for people from poor backgrounds. The same has been the case in England and America in the not-so-distant past. However success in the army is not automatic for everyone and the fact is that there are many poor soldiers, who have to live on their salary and access to rations at subsidised prices (if the rations have not been sent elsewhere for other purposes). In addition it is very common to find soldiers, even the poorest, having more than one wife and family in different parts of the country. Typically these families are large, with little spacing between births. The poverty of army families is recognised by many Somalis as a commonplace, as it is also recognised that for some the army still has its opportunities as well. For the families of such men, income-generation assistance to the wives would seem to be the only possible option. This could be well justified. One must wonder under what circumstances and for what reasons many of these women have become married to members of a group whose poverty and difficult living conditions are so well known. B. Other government employment: In both groups a number (8) of men were employed by the Government. In the case of the four employed by the Local Government the type of work they were doing is not known. It could of course range from low-paying positions such as watchman or cleaner, to very remunerative work such as a market tax collector. The one man known to be working as a government watchman (Hospital) was very poor. Apart from his salary and the dhacdhac (informal fringe benefits) associated with his work, the family's only other income were refugee rations. The two youngest children were both significantly malnourished. The three other positions appear to be well paid but the case of the man working in the Central Bank illustrates the difficulty of interpreting family economic circumstances on the basis of occupation alone. In this case there were 13 members in the household. Seven of these had come from Hargeysa in late 1988, including one soldier, disabled as a result of the fighting there. The father was the only one employed but the family was receiving support from relatives overseas. For men working in low-paying government positions the only appropriate income-generation strategy would be to seek alternative or second occupations, as was the case with Hassan, the hand-cart trader described in Section 4.2.3.4.1. C. Animal traders and brokers: There was a marked difference in the significance of casual labouring work in the two groups. In Yaqshiid this was the single most important source of employment, whereas in Belet Weyne only one person was so employed. In contrast, in Belet Weyne, trading and brokering sales of animals was a major activity. Just as casual building labouring can lead to work as a fuundi (mason) in Mogadishu, the same relationship appears to be the case with animal brokering leading to animal trading in Belet Weyne. Both occupations are also characterised by unpredictability of earnings from day to day, but perhaps more so in the case of labouring work. Although animal brokering appears to be a relatively easily-available casual income-earning opportunity, requiring minimal capital, examination of the previous work of those who were brokers suggests that brokering was something resorted to rather than aspired to. This could equally well be said of casual labouring work in Mogadishu. With this occupation, in contrast to labouring work, it could be quite easy to extend credit to enable men to move from brokering to trading, and thus gain a larger profit margin. However it would be appropriate to first examine why individual brokers had not already been able to move up to the level of traders. Was lack of access to credit the problem, or were there other factors involved ? D. Private-sector wage employment: It seems that plans for income- generation assistance to the poor frequently assume that most of the expected beneficiaries will either be self-employed or seeking self-employment. However, in the Yaqshiid group, 60% (18) of the men were employed by other people, and in Belet Weyne 53% (22). Apart from the armed forces, the employment opportunities taken up by many of the poorest and least educated frequently include labouring and work as watchmen. Assistance planning for this group needs to take account of at least three issues: 1. Labouring work is usually irregular work. The unreliability of income from this work is at least as much of a problem as the level of the wage itself. 2. Those on fixed wages are in a far more vulnerable situation than self-employed traders during times of inflation. In most cases, traders are able to make continual adjustments to their selling prices, in response to changing costs. Labourers do not have the same degree of control over the value of their services. 3. Movement between casual wage employment, particularly labouring work and trading occupations, is not completely free. Many men, especially those from northern and nomadic backgrounds, look down on labouring work and avoid it where possible. On the other hand, certain southern Somalis who are prepared to seek labouring work will not normally consider small-scale trading as their next option, but will look instead for other wage-employment opportunities. Given the fact that the pool of casual labour is continually being replenished by migrants and from population growth within the city, the best aim for those targeting assistance to the casual labour group, may be to facilitate movement into new occupations for men from the most vulnerable families (defined in nutritional terms). One means would be to provide subsidised apprenticeships to enable labourers to learn masonry and mechanical skills more rapidly, which would give them access to better paid positions in areas which would be likely to be within their aspirations. The limitation of such an approach is that it is a very individualising solution which might simply tend to develop a dependent social relationship, between apprentice and "master" (see Section 5.3.3. for further comment). An alternative approach, of more medium-term impact might be to encourage or reinforce mutual self-help between labourers, through the formation of labourers' groups on a very simple and informal scale. Informal mutual self-help activities already occur in other small businesses such as dulaal (brokers)( interviewed in Belet Weyne), or with abakaar (milk traders) (interviewed in Mogadishu). One possible group initiative might be a very simple unemployment insurance scheme, whereby daily contributions by members were accumulated and used to pay members during periods of illness (up to an agreed limit). This would provide some degree of income security, which is as important for poor families as the absolute amount of income they receive. Careful investigation would be needed before any specific measures were taken. It may be that there is high occupational mobility which would make any organisations very unstable. In addition the nature of existing forms of cooperation amongst labourers should be identified, for example group labour contracting. If there are few, or none, the reasons why need to be carefully considered. In the background, there are already large scale cooperatives which exist on a formal basis. Although these are of limited benefit to ordinary labourers, for example those working in the quarries, they would have a vested interest in co-opting any new group activities involving those traditionally within their constituency. To the extent that any informal labourers' groups were formed primarily as as response to their families relationship with the local MCH, it could well be argued that they could and should be facilitated independently of the national Cooperative structures. E. Trading. Comments on small trading activities (e.g clothes selling) are made in the section of this report dealing with the results of the survey of small businesses in Belet Weyne. 3.2.3.3 Current and previous occupations of fathers and mothers (cont) .1 Mothers in YAQSHIID MCH SAMPLE (39) %%%%%,%%%%%%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%%%% % % Current % Past % Reason for change % %%%%%4%%%%%%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%%%$% % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % % % none domestic servant she had children % % none student % % none farmer husband came to Mog % % none clothes washer % % none cigarette seller sickness % % none selling khat made illegal % % % % mat maker none % % mat maker none % % mat maker none % % mat maker none % % mat maker farmer husband died % % mat and basket maker maize seller relatives children % % came to stay % % mat maker domestic servant empl'er left country% % % % muufo, mat maker none % % cooks popcorn for son none % % to sell % % sells oil, sugar domestic servant empl'er left country% % sells vegetables none % % sells meat none % % sells vegetables none % % sells "things in market" none % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% .2 Mothers in BELET WEYNE MCH SAMPLE (44) %%%%%,%%%%%%%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%%%%% % % Current Employment % Past Employment % Reason for change % %%%%%4%%%%%%%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%%%%$% % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none none % % none mat making no money for leaf supply % % none mat making sickness % % none tea-shop came to BW as a refugee % % none tea shop to get medical treatment % % none sold petrol fighting in Hargeysa % % none teacher more children to care for% % % % mat making none % % mat making none % % mat making none % % mat making,washes clothesnone % % cows in house compound none % % farming / repair cariish none % % repairs cariish none % % meat seller none % % meat seller none % % sell tea,rice,canjeero none % % sell tobacco, rice,sugar none % % clothes seller none % % soldier (health) none % % teacher, Min of Ed student graduated % % unknown, Min of Finance none % % working but unknown none % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Observations: The proportion of mothers who reported earning income were the same in both YAQSHIID (36%) and in BELET WEYNE (36%). Data on the proportions of all women working are available for three areas in Mogadishu (Davies,1986) for comparison: Yaqshiid squatter area 29%, and Waaberi (an average income area of Mogadishu) 22%. As with male unemployment, female unemployment seems less common in the poorest areas and groups. The key questions concerning women who are not earning any income at present is what their reasons are for not doing so, and what are their attitudes towards employment in the future. These questions are discussed in detail in Section 3.2.4.1 of this report. Those women who were earning income were involved in four basic types of activities: .1 Home-based productive activities: making mats, baskets, muufo, popcorn. .2 Selling various items in the markets: vegetables, meat, oil, sugar, tea, rice, canjeero, tobacco, soap and clothes. .3 Working in the employment of others, in the private sector: salaaxid/malaas (cariish repairers) and clothes washers. .4 Working for the government: teachers, military For each of these types of activities consideration can be given to the most appropriate strategy for providing income-generation assistance. Some broad options are outlined below, for each of the activity types: A. Home-based production for sale. There is already a reasonable range of goods produced in homes, for sale to neighbours or in the local markets, both in Mogadishu, and in Belet Weyne. Activities seen in the course of this research have included: mat, basket, and broom making, muufo and caajin making, popcorn, roasted peanuts, boiled sweets and sinsin (sesame sweets), eggs and milk, "cool drinks", and sambusa. Many have the advantage of being consistent with other home-based activities women are responsible for (child care, cooking and cleaning). The limitations include the need for other people to help sell them if the quantities involved are very substantial (peanuts and sweets often have to be sold by the children of the producers) or if the quantities involved are too small to justify a trip to the market (women with chickens in their compounds sold their eggs to market traders, who specialised in selling eggs and chickens). The most useful contribution would be to help increase the diversity of such products on sale. This might help mitigate the squeeze on profit margins resulting from too many producers relying on the production of one or two items for sale such as mats. Lessons could be learned from South-East Asia where the range of home-produced goods, especially food items, is far greater (See Equity Policy Centre, 1984). Are there also lessons to be learned from within Somalia? If so, how can the potential beneficiaries themselves be involved in identifying them?. How can those skills which are well developed in one part of the country be transferred to another part? B. Retail trading. In the two groups the range of goods women were retailing outside their homes was relatively limited. However, in the markets as a whole, both in Mogadishu and Belet Weyne, it is plainly evident that women are involved in selling a wide range of goods. Assistance strategies in this area could include: - helping women start up in such activities; - helping increase the profits of those already involved in trading; - helping women accumulate savings from these profits; - helping women make the transition from one occupation to another where there are clear limitations to their earnings in their current occupation, or when they see opportunities in other areas. Some practical comments on these approaches will be made in section 4 of this report, which reports the results of a survey of low income occupations. C. Wage employment in the private sector. In contrast to the men in the sample there appeared to be almost no women employed in the private sector (is clothes washing and repairing the mud walls of cariish considered as wage employment?). In the Yaqshiid sample three women reported that they had previously worked as domestic cleaners, but none were doing so at present. Why is it the case that the proportion of women employed in the private sector is so low?. Partly it may be due to the very low literacy rates amongst the women interviewed. One reason why domestic work is attractive to some women is that no literacy and numeracy is required. However, in other areas such as transport, manufacturing and construction a widely entrenched gender bias, held by both sexes, against women's involvement is undoubtedly a major factor. Even those women who want access to such occupations face major problems of how to find the opportunities to learn the basic skills required. The basic question then in this situation is how opportunities for skills training in male dominated areas can be opened up to women. Rather than struggling forward in any direction it might be useful to take a historical perspective. Changes in attitudes to occupations happen over a period of time. What occupations in Somalia have moved from being male to female dominated?. What occupations have become more male dominated?. Understanding these changes might help identify a strategy for choosing which types of occupational training to focus efforts on to bring about increased access for women. D. Wage employment in the public sector. Although only a very small number of women (3 in the Belet Weyne sample) were employed by government an appropriate strategy in this area could still be relevant. In the current economic environment, and given the emphasis on reducing the size of the public sector workforce in the current structural adjustment program, the most appropriate strategy would be to focus on defending current levels of female employment in the public service, in absolute and relative terms. Since women occupy a disproportionate number of the lowest echelon positions it is likely that they will be particularly vulnerable to large-scale retrenchment. Documentation of current employment levels of women in the public service, monitoring changes and promotion of retrenchment policies which do not discriminate against women would be a useful strategy for organisations specially concerned with women's issues. This could be done in regional towns as well as the capital city. 3.2.3.4. Occupational mobility .1 Fathers. Because information on the family was gathered through interviews of mothers at MCH centres, information on the past occupations of the fathers is not complete. Almost 43% of the women interviewed did not know what type of work their husbands had done before their current occupation. It may be that this reflects two processes. On the one hand the men concerned have not changed occupations during the marriage. This suggests a fairly low rate of occupational mobility. The other possible reason is that the men have not shared much information at all with their wives about their lives before they were married. From the information that is available in the above tables, the following types of reasons for changes in occupations can be identified. %%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Number of cases% Reasons for changing occupation % %%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$% % Negative % % 9 sickness and old age % % 5 displaced by fighting/conscription/fire % % 3 previous employment was terminated % % 2 loss of working capital used for previous business % % % %Sub-total 19 % % Positive % % 5 insufficient income in previous occupation % % 4 attraction to another occupation e.g soldier % % 3 moved voluntarily from one place to another % % 2 promotion/transfer % % 2 desire to leave past occupation e.g the bush % % % %Sub-total 16 % % % %Total 35 % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% .2 Mothers. Sixteen of the 40 women had reported changing their occupation in the past. While there may have been some under reporting of changes, the general impression is that the amount of occupational mobility in these two groups of women was quite low. The most common reasons for changing are summarised below: (See Table next page) %%%%%%%%%%%%%%%%,%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% %Number of cases% Reasons for changing occupation % %%%%%%%%%%%%%%%%4%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%$% % 4 sickness of mother or child, or death of father % % 3 additional children in the house (childbirth or % % relatives) % % 3 employment finished % % 2 displaced by fighting (Ethiopia/northern Somalia) % % 1 moved voluntarily from another place, with husband % % 1 not enough money to buy supplies % % 1 graduated % % % %Total 15 % %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% Only in one case does there seem to be anything like upward mobility in occupation. Fortuitous changes in circumstances seemed to be a much more common reason for changes in occupations. Although the number of cases being compared is very small it seems as though the men had more control over the circumstances in which they changed occupations. Both the male and female groups do have one major cause in common. Sickness was the most common reason for occupational change in both groups. This supports the relevance of the proposal already made concerning a simple contributory unemployment insurance scheme, for daily wage labourers, but also for others who depend on casual employment. It also suggest that assistance with the health problems of such families may also be an appropriate form of income generation assistance in itself. 3.2.4 Women's attitudes to employment: present and future The focus of the following section is on women's attitudes towards employment. An assumption has been made that if women are earning income, they are more likely than men to use a significant amount of their current or extra income to meet the needs of their children. Typically one of the arguments raised against promoting increased employment opportunities for women is that less time will then be available for their others duties e.g child care, and the children will suffer as a result. Studies of breast feeding practices, one area where employment might be thought to have an effect, have in fact shown that factors other than employment are far more important. Regardless of the channel of effect, if women's employment was a major causative factor it would be expected that a very large proportion of women with malnourished children would be earning income (and most likely be from all income classes). This is not the case. It seems more plausible to assume that in most cases both mothers employment and child malnutrition are both consequences of family poverty, not one the cause of the other. During the interviews women were asked the following five questions: 1. What is the reason you are not working (for money) at the moment? 2. When you have to leave your house for a long period who looks after your children? 3. If you find work in the future what type would you like