Resource needs to support orphans and vulnerable children in sub-Saharan Africa

Abstract

In sub-Saharan Africa, 43 million children under the age of 18 have lost one or both parents to AIDS, conflict or other causes. This large number strains systems by which families and communities have traditionally provided care for orphans. Support for some orphans is being provided by a variety of government, community and non-governmental organizations but this assistance reaches only a small percentage of those who need it. This paper estimates the funding required for necessary support to those most in need. We estimate that US$1–4 billion will be required annually by 2010, depending on whether support is provided to all orphans living below the poverty line or just those in most need. This is at least four times current funding and should be a priority topic for donor and national government resource allocation decisions this year.

Orphans, AIDS, resource needs

KEY MESSAGES

• 43 million children in sub-Saharan Africa have lost one or both parents. The magnitude of the problem has strained traditional support systems, placing many children in critical need of support from outside their extended family.

• About US$1.7 billion (US$1–4 billion) will be needed annually by 2010 to provide essential support services to all orphans and vulnerable children who need them in sub-Saharan Africa.

Introduction

Over 140 million children under the age of 18 in the developing world have lost one or both of their parents. In sub-Saharan Africa alone there are 43 million orphans, representing more than 12% of all children (UNAIDS et al. 2004). Although most orphans are cared for by family members or communities in some way, many of these families are living in poverty. Some form of public assistance is required to provide these children with adequate food, health care, clothing, education and psychosocial support. The need for public assistance varies by country depending on the number of orphans, the socio-economic conditions and local decisions about the type of support to provide and the best way to provide this support.

The amount of resources available for HIV/AIDS programmes has increased remarkably in the last few years. Many donors are willing to expand support for programmes to assist orphans and vulnerable children, but this expansion is hampered by the lack of good estimates of the amount of support needed. We undertook this effort to provide a sounder basis for estimates of resources needed, and to provide information and methods to support national efforts to estimate the resources required to expand programmes.

Resources needed are estimated as the number of children needing support multiplied by the coverage (the percentage receiving support) and multiplied by the unit cost of providing the service. Estimates are prepared for different types of support including education, nutrition, health care, family/home, economic stability, community support and organizational costs. Children also need shelter but we have assumed that most households can most easily provide shelter and that programmes should focus on support to families to care for orphans in their homes rather than establishing separate orphanages. We have prepared estimates for 43 countries in sub-Saharan Africa with the most orphans for the period 2004–2010.

Research methodology

Estimates of resources needed for orphan support require information on the number of orphans and the cost. Projections of the number of orphans by type (maternal, paternal and double) have been made by UNAIDS using the Spectrum software (Stover 2004). A detailed description of the methodology is provided elsewhere (Grassly and Timeaus 2005). These projections are based on estimates of adult HIV prevalence and assumptions about the progression of HIV infection to death provided by UNAIDS, and demographic data provided by the United Nations Population Division (United Nations Population Division 2005).

The costs of supporting orphans with specific services in some countries have been reported elsewhere (Boerma and Bennett 1997; Drew et al. 1998; Desmond and Gow 2000; McGreevey et al. 2004) and range from US$234 to US$702 per child per year. These reports cover a limited number of countries and services. In order to obtain a more thorough understanding of the costs of support in sub-Saharan Africa, we collected unit costs in 22 countries (representing 70% of all orphans in sub-Saharan Africa). We hired local consultants to administer a questionnaire to key organizations providing orphan support. Organizations were identified through discussions with the UNICEF office in each country and with the largest donors for programmes for orphan and vulnerable children. Organizations were asked to provide direct unit costs for each service individually plus the indirect or overhead rate that covers administrative and other non-allocable costs. Most organizations had this information as they needed it for grant proposals.

In all, 300 organizations provided information from 7400 sites supporting nearly 2 million children. Organizations reported costs only for the services they provided, which usually did not include the complete list of all services examined in this report. The costs are financial costs estimated from their expenditure and used for budgeting. The unit cost for each country in the survey was calculated as the median unit cost across the organizations that reported a cost for that service. Some services are free to the organization supporting the child if the government provides the service free of charge. Although there is a cost to the government to provide these services, we did not include these costs here since the government services support all children, not just orphans. Generally the cost of supporting orphans is a small portion of the total cost of supporting all children.

Number in need of support

The number of orphans and vulnerable children in need of public support is not known precisely. UNAIDS, UNICEF and USAID have jointly prepared estimates of the number of orphans by country and type for 2003 to 2010 (UNAIDS et al. 2004). However, not all of these children need public support. Some may be living well with their surviving parent while others may be living with relatives who can afford to support them. Our definition of orphans in need includes only those living in households with incomes below the national poverty line (World Bank 2002). Of course, many children who are not orphaned are also in immediate need of assistance. The identification of children in need of assistance can best be done through community assessments. Many countries have adopted their own definitions of orphans that differ in important ways from the standard definition used by UNICEF and other international organizations (a child under the age of 18 who has lost at least one parent) (POLICY Project 2004).

After limiting the target population to orphans living under the poverty line, for the purposes of this analysis we use three further definitions of need as proxies for these community assessments. They are double orphans (children under the age of 18 who have lost both parents), double and near orphans (‘near orphans’ refers to children with a chronically ill parent who will become orphans in the next year), and double orphans and near orphans plus one-half of maternal and paternal orphans. These definitions refer to all orphans not just those children orphaned because of AIDS. (These estimates are based on current levels of anti-retroviral treatment (ART). If the number of people treated expands very rapidly in the next few years, fewer parents will die and the number of orphans would be about 5% less.) Near orphans are especially vulnerable since they may receive inadequate care if they have a chronically ill parent and they will shortly become new orphans, requiring significant adjustments in their living arrangements. Community workers can identify families with chronically ill adults. In our projections we estimate this number as the children who will become single orphans the following year.

These definitions are intended to serve as proxies for orphans and vulnerable children in need of public support and are used only for the purposes of estimating resources needed. It is not intended to suggest that programmes use these rules to identify children in need of support.

The total number of orphans is projected to grow from 21.6 million in 2005 to 23.5 million in 2010. In 2010 there will be an estimated 6.4 million maternal orphans, 8.4 million paternal orphans and 4.4 million double orphans, plus an additional 2.5 million children would be orphaned in the coming year (near orphans).

Using the definitions of need presented above, the number of children in need of public assistance grows from 2003 to 2010 as follows: 3.6 to 4.4 million double orphans, 6.1 to 6.9 million double orphans and near orphans, and 16.3 to 18.7 million double, near, maternal and paternal orphans. The distribution by age is heavily weighted towards older children, since they are composed of children newly orphaned and those orphaned some years ago at younger ages who have subsequently aged into their teenage years. For example, the distribution of single and double orphans living below the poverty line in 2010 is 400 000 aged 0–4, 2.2 million aged 5–11 and 4.2 million aged 12–17.

Essential services

Children need various types of support ranging from those things necessary for survival, such as food and health care, to those interventions that will provide a better quality of life in the future such as education, psychosocial support and economic self-sufficiency. In an ideal world all children would have access to all types of high quality services. In the real world many children, orphaned and not, are malnourished, sick and without shelter. Some argue that a comprehensive programme to support children should include all essential elements including food, health care, education, clothes, shoes, bedding, psychosocial support, economic self-sufficiency, etc. Others hold that some of these elements are not ‘essential’ or far exceed the situation of most children living in poor households with both their parents.

Each country will make its own decisions about what types of support to provide in light of the availability of funding, level of need and socio-economic situation. In this analysis we provide estimates of resources needed for six categories of support:

• Food: Food and clean water are the most basic need for all children. Food supplied from external sources could actually reduce food security in the long run if it disrupts the local market, but food procured locally or produced through community gardens can contribute to local food security. Food may be provided as either bulk grain needing preparation or as cooked meals.

• Health care: The need for health care includes childhood immunizations and vitamin supplements for children under five, routine health care for all and reproductive health services for older children aged 10–17. In some countries health care is free for all children or for the youngest children. However, patients often have to pay for drugs and supplies. Some have argued that orphan-related programmes should advocate for free health care for all children rather than focus on providing funding to pay for care for children. However, in that case additional resources would need to be made available to governments to provide free care to families.

• Education: This includes school fees where they exist, funds required for uniforms, books and other supplies, and special fees. Many countries have eliminated school fees and additional advocacy efforts could help to eliminate them in other countries as well, but the extra costs of uniforms, supplies and special assessments can still be substantial.

• Family/home support: This category includes clothes, shoes, bednets and economic self-sufficiency. The need for bednets will vary depending on local climate and other conditions. In many cases donated clothes and shoes are available at no cost, but reliance on donated goods may not be sustainable as programmes scale up considerably. Economic self-sufficiency refers to programmes to provide older children and/or their families with economic support such as microfinance loans, skills training, grants or seeds.

• Community support: This includes identification of vulnerable children and funding for community workers who can assess needs, organize support and provide some counselling and individual support. Many community workers will be volunteers but significant funds may still be required for training and transportation.

• Other services: We have not explicitly included costs for national-level advocacy and legislative reform. These may be needed in many countries to address specific issues such as school fees, the cost of health care or child protection. Also many important activities such as memory books, camps etc. are assumed to be covered under community support but these may require additional resources beyond the community worker training and support included here. We have included an estimate of the costs of administering support programmes including fund raising, planning, research etc. This category applies to the organizational costs of non-governmental organizations (NGOs) providing support or central administration costs for government-run programmes.

Coverage

The coverage of essential support services is assumed to increase from its current level in 2003 to 100% of those in need by 2010. Estimates of current coverage are based on several sources, including (1) estimates of children currently served that have been collected from reports of various organizations providing orphan support (Begala 2004), (2) the number of children served as reported to us by selected organizations as part of a costing survey described below, and (3) information from a 2003 survey on the coverage of essential services for HIV/AIDS (POLICY Project 2004). Based on this information, the amount of resources currently spent on orphan support and our definition of the number of children needing support, we assume that regional average coverage rates in 2003 are roughly as follows: 11% for primary schooling, 3% for secondary schooling, 4% for nutritional support, 8% for health care, 6% for family/home support and 1% for economic self-sufficiency and community support.

The goal of 100% coverage of the target population by 2010 is certainly ambitious. But it is consistent with international goals for HIV prevention, which call for comprehensive prevention programmes in all countries by 2010, and for anti-retroviral treatment, with the goal of universal access to treatment by 2010.

Unit costs

The median unit costs across all countries as determined by our survey are shown in Table 1. Appendix 1 (available as Supplementary Data at Health Policy and Planning Online) gives the detail of the median costs by country and type of service as well as the number of organizations reporting and the number of sites included. The typical costs per child are US$480 for children aged 0–4, US$690 for children 5–9 and US$830 for children 10–17. These costs are quite high since the cost of a full package of services exceeds the per capita income of many poor families. In many cases the costs are high because many organizations are supporting a small number of children in each site. They do not have resources for bulk purchase and often have to pay full commercial prices for items they purchase, and, in some cases, use private sector health care providers because public facilities are not accessible.

Table 1

Median costs per child per year in US dollars for sub-Saharan Africa and projected costs in 2010 if programmes reach full scale

	2004 (US$)	2010 (US$)
Education
Primary school
(1) School fees	18	18
(2) Uniforms	14	14
(3) Books and supplies	17	17
(4) Special fees/assessments	10	10
Secondary school
(1) School fees	103	103
(2) Uniforms	21	21
(3) Books and supplies	44	44
(4) Special fees/assessments	33	33
(5) Skills training	98	98
Nutritional support
(1) Food	310	69
Health care support
Child 0–4
(1) Full course of childhood immunizations	1	1
(2) Vitamin A, zinc and iron supplements	12	7
(3) Routine health care	27	3
Child 5–9
(1) Routine health care	32	3
Child 10–17
(2) Routine health care	28	3
(3) RH and HIV prevention information and services	41	4
Family/home support
(1) Bednet	7	2
(2) Clothes	28	7
(3) Shoes	16	7
(4) Blanket and bedding	16	2
Economic stability support
(5) Microfinance	110	110
(6) Income generating activities	86	86
(7) Seed for food crops	35	35
(8) Self-support grants	93	93
Community support
(1) Identification of children in need in the community	10	10
(2) Outreach for street children	126	2
(3) Train and support one full-time community worker	3	2
Organization costs
(1) Administration, monitoring & evaluation, fund raising, reporting as a percentage of all costs	16%	16%

The unit costs vary with the quality and intensity of services. For example, in Zimbabwe the cost of providing the grain required for a simple porridge meal to school-children 200 days a year is only US$11 a year, while providing dry food for 2100 calories per day can cost up to US$75 per person per year in Zimbabwe (Stevens et al. 2004). Prepared meals can cost much more.

We have assumed that unit costs for some services will decline as programmes expand due to either economies of scale or changes in the way services are provided. We have assumed that economies of scale apply to food, blankets and bedding, and bednets. As programmes expand it should be possible to purchase items in greater bulk at lower prices. We used the World Food Programme estimates for the cost of food at full scale, which is just the cost for bulk dry food and excludes preparation costs, and used the lowest national costs provided to us in our survey for most other services.

We also assume that changes in approach will result in lower costs for health care and community support. For health care for older children and identification of children in the community we reduced costs at full scale to one-tenth of current costs. Our assumption is that organizations would make greater efforts to utilize free health care services and volunteer labour if their programmes were much larger. For community workers we assumed that the unit costs could be reduced somewhat through greater use of volunteers, but that funds would still be needed for training, recruitment and transportation. For all services the unit costs are assumed to decline from their present values in 2003 to these new values by 2010 as the scale of the programme expands. This assumption reduces the typical cost per child supported from US$652 per child in 2004 to US$224 per child in 2010.

An alternative to funding each type of support required is to provide conditional cash grants to families or communities. Several countries are investigating such mechanisms. While small monthly grants can improve conditions significantly, they will not provide the same quality of support as a more comprehensive programme. It may be that providing larger conditional grants to families or communities would be a more efficient way of providing support, but many issues remain to be resolved, such as identification of those who need support and ensuring that funds are not wasted.

Results

The costs for a full package of services for children most in need in sub-Saharan Africa will grow to between US$1 billion and US$4 billion annually by 2010, depending on how many children are served. Table 2 and Figure 1 show the estimated need under several different scenarios in 2010. Table 3 provides more detail about the distribution of needs by type of service for the scenario assuming support to all double and new orphans living below the poverty line.

• Food support requires the most resources, about 32% in 2010, even with the assumption that the average costs per child per day can be reduced to US$0.19.

• Education costs are the next largest item. This projection assumes that primary school fees continue in those countries that have them today. Eliminating fees for primary school would reduce the resources needed for education by 11%. Secondary school fees account for 19% of the education total and skills training accounts for 18%.

• Family and community support accounts for about 20% of total resource needs. Clothes, shoes and bedding account for about 8%, with programmes for economic self-sufficiency accounting for most of the rest.

• Health care costs are relatively small compared with other services, partially because services are free in some countries. When services are free there is still a cost to someone, in this case the national government. But since the government provides these services to all children, we do not count them as resources spent on orphan support. We assume that the unit costs of most routine health care services decline by 2010. The largest costs are for reproductive health services for older children and vitamin supplements and immunizations for children under five.

• Community support is the smallest component in 2010. This assumes that most community services are provided by volunteers. Funding is required for recruitment, training and transportation.

• Organizational costs account for about 16% of the total. It is unclear whether this figure could be reduced as programmes reach greater scale or whether the challenges of identifying children in need and fairly administering large programmes would make it difficult to reduce organizational costs.

Table 2

Resources required in 2010 to support all orphans and vulnerable children in need of public support by type of support with three definitions of ‘need’, in millions of 2004 US dollars

Type of support	Double orphans (US$ millions)	Double orphans plus near orphans (US$ millions)	Double orphans, near orphans and 50% of single orphans (US$ millions)
Food	340	530	1400
Heath care	35	57	160
Education	300	460	1200
Family/home support	220	330	840
Community support	46	46	46
Organizational costs	180	270	680
Total	1100	1700	4400

Table 3

Resources needed in 2010 by type of support to support double orphans and near orphans in sub-Saharan Africa

Type of support	Millions of US$
Education	458
Primary school
(1) School fees	42
(2) Uniforms	40
(3) Books and supplies	50
(4) Special fees/assessments	23
Secondary school
(1) School fees	98
(2) Uniforms	28
(3) Books and supplies	49
(4) Special fees/assessments	36
(5) Skills training	92
Health care support	57
Child 0–4
(1) Full course of childhood immunizations	1
(2) Vitamin A, zinc and iron supplements	3
(3) Routine health care	1
Child 5–9
(1) Routine health care	9
Child 10–18
(1) Routine health care	19
(2) RH and HIV prevention information and services	25
Food	546
Family/home support	333
(1) Bednet	1
(2) Clothes	58
(3) Shoes	54
(4) Blanket and bedding	15
(5) Microfinance	62
(6) Income generating activities	63
(7) Seed for food crops	31
(8) Self-support grants	50
Community support	46
(1) Identification of children in need	5
(2) Outreach for street children	1
(3) Community worker	40
Organization costs	271
(1) Administration, monitoring & evaluation, fund raising, reporting	271
Total	1700

Figure 1

Resource needs in 2010 by type of support under three definitions of need, in millions of US dollars

Discussion

These estimates indicate that about US$1.7 billion (US$1 billion to US$4 billion) are needed annually by 2010 to address the needs of orphans and vulnerable children in sub-Saharan Africa. The estimates would be less if countries chose to focus public support on only some parts of the support package used here or more if it is not possible to reduce the unit costs of support for food and health care as we assume.

The major difference between the three scenarios shown here is the definition of the population in need of support. In all three scenarios we focus just on children living in families below the poverty line. Those children most in need are those who have lost both parents (double orphans) and new orphans who need to find a new living situation quickly. Focusing efforts just on these children will require US$1.1–1.7 billion annually by 2010. But that excludes many single orphans who may also need assistance because the other parent is absent or unable to carry the full burden of supporting the children alone. Expanding the definition of those in need to include half of single orphans more than doubles the resources needed.

Although current funding for support is not known, an analysis of reported funding by the largest support organizations in the 17 most affected countries in sub-Saharan Africa suggests that funding in 2003 was in the order of US$200–300 million. Thus the need in 2010 represents at least a quadrupling of resources. The cumulative funding required from 2005 to 2010 ranges from US$7–28 billion.

These estimates refer to resources required from all sources except household sources. That is, they include resources needed from national governments, bi-lateral and multi-lateral donors, foundations, commercial sector organizations and individual donors.

We recognize that the average cost per child is high compared with the per capita income of countries in sub-Saharan Africa. We also recognize that orphans are receiving some of these services now. For example, many are in school and all are receiving some food, as caregivers are doing what they can to provide the best care possible with their limited resources. We really need to know the additional resources required to provide adequate care to all children. For these reasons we have limited our estimates to children living below the poverty line, as they are the least likely to have adequate support.

These estimates are based on the current unit costs of providing services. They do not include resources that may be needed for infrastructure expansion to meet the additional needs. On average these costs are not expected to be large, since meeting these targets would result in only a 6% increase in the numbers of all children (orphans and non-orphans) receiving services. However, in some countries the infrastructure needs may be substantially higher.

There are no internationally agreed, upon standards that define which children most need public support or the essential package of services they should receive. Approaches to providing this support and costs will vary from country to country. In this analysis we have tried to show the range of resource requirements by providing estimates for three different definitions of need and five different categories of support. These categories are not intended to be recommendations for programming, but rather to serve as useful proxies for the number of children in need and the types of support that would be identified by communities and provided by them if adequate funding were available to them.

Although these estimates represent a large increase in resources over current levels, the cost of doing nothing would be even higher. More than 15% of children are orphaned in at least 11 countries in sub-Saharan Africa and overall there are over 40 million orphans (UNAIDS et al. 2004). Failure to address the needs of these children would not only seriously jeopardize their welfare, but could lead to significant social disruption and economic loss (Bell et al. 2003).

These resource needs estimates have been included in the UNAIDS estimates of the resources needed for the global response to the HIV/AIDS epidemic (UNAIDS 2005). They show that orphan support represents about 12% of the need for all HIV/AIDS prevention, treatment and mitigation needs.

About 15 countries in sub-Saharan Africa have recently developed National Orphan and Vulnerable Children Action Plans. In most of these countries the approach described here was used to prepare plan budgets, and the unit costs collected through our survey have been used for costing the plan. Efforts are underway to develop costed action plans in all countries in sub-Saharan Africa. Once these are all completed they can serve as the best estimate of resources needed to fund national efforts. In the meantime, the estimates presented here represent our best estimates of the resources needed to support orphans and vulnerable children in sub-Saharan Africa with a comprehensive set of services.

Acknowledgements

The authors would like to thank the many people who provided input into the process of developing this work. These include Alex Yuster, Doug Webb and Mark Connolly (UNICEF), Ron Waldman (Columbia University), John Novak, Linda Sussman and John Williamson (USAID), Marito Garcia and Mehahem Pryres (World Bank), Sara Siever (Francois Xavier Bagnoud), Steven Forsythe (Instituto Nacional de Salud Publica), Jane Begala (Constella Futures) and Leanne Dougherty (Abt Associates).

Supplementary Data

Appendix I is available as supplementary data at Health Policy and Planning Online