CARE officially adopted household livelihood security (HLS) as a programming framework in 1994. Over the past five years, CARE has been working to institutionalize the approach in its programming worldwide. This has been neither a smooth nor an easy process. Significant progress has been made in improving concepts, strengthening their application and understanding their implications on programme design and evaluation.
This paper describes how HLS has been operationalized in CARE. Drawing on lessons learned from a number of countries, the paper shows how livelihood concepts and tools have been taken into account in strategic planning, diagnosis, design, implementation, monitoring, reformulation and evaluation.
Household livelihood security continues to be the cornerstone framework that CARE uses to carry out its programming efforts. It allows CARE to have a more holistic view of the world to inform its programming decisions, enabling the organization to understand better the root causes of poverty. In addition, it helps clearly identify opportunities and leverage points for positive change. Application of the livelihood framework should not be considered a linear process but rather a flexible, dynamic and iterative process over time.
Taking a holistic view does not always mean that one must undertake multiple interventions. Application of the HLS framework can be done using various entry points.
Over the past several years, CARE has identified several analytical lenses that have been incorporated into an HLS holistic analysis to understand better the root causes of poverty. These analytical lenses include basic needs, a human rights perspective, civil participation and action, gender and the policy environment. These various lenses are significantly influencing the future directions of CARE programming.
In the end, the HLS framework is helping CARE make strategic choices about where to concentrate its limited resources and how to leverage its comparative advantages to achieve the most positive and lasting change. It is through these efforts that CARE will contribute to the global effort to end poverty.
SECTION I. INTRODUCTION
A brief history of the inception of household livelihood security
Definition of livelihood security and the underlying principles
Putting theory into practice - dissemination and decentralization
SECTION II. OPERATIONALIZING THE HOUSEHOLD LIVELIHOOD SECURITY FRAMEWORK
Creating livelihood security profiles
Working with partners
Strengthening civil society
Long-range strategic planning
Diagnosis leading to design.
The project design framework
Developing focused project and programme strategies
Using a livelihoods framework to redesign existing projects
Reflective practice: achieving greater gender equity
SECTION III. CONCLUSIONS AND FUTURE DIRECTIONS
APPENDIX 1. IMPLEMENTING HLS ACROSS THE RELIEF-TO-DEVELOPMENT CONTINUUM
Livelihood promotion (sustainable development)
Livelihood protection (mitigation)
Livelihood provisioning (emergengy response)
Livelihood recovery (rehabilitation)
APPENDIX 2. USING A LIVELIHOOD FRAMEWORK TO INFORM POLICY
APPENDIX 3. DIAGNOSTIC SEQUENCING AND METHODS
APPENDIX 4. DEVELOPING FOCUSED PROJECT STRATEGIES: THE MALAWI CASE STUDY
Malawi's central regional programme
APPENDIX 5. REFLECTIVE PRACTICE: THE ZAMBIA CASE STUDY
The Livingstone food security project
Household livelihood security (HLS) has become CARE's basic framework for programme analysis, design, monitoring and evaluation. HLS grows out of a food-security perspective but is based on the observation that food is only one important basic need among several, and adequate food consumption may be sacrificed for other important needs. Given that the causes of poverty are complex, HLS provides a framework for analysing and understanding the web of poverty and people's mechanisms for dealing with it.
CARE officially adopted household livelihood security as a programming framework in 1994. Over the past five years, the agency has been working to institutionalize a livelihood approach in its programming worldwide. This has been neither a smooth nor an easy process. CARE has put a lot of effort into mainstreaming HLS by developing tools and methods, training staff, encouraging reflections and learning how to improve the framework. Significant progress has been made in improving the concepts, strengthening their application and understanding their implications on programme design and evaluation.
The purpose of this paper is to describe how the household livelihood security framework has been operationalized in CARE. Drawing on lessons learned in a number of countries, the paper attempts to show how livelihood concepts and tools have been taken into account in diagnosis, design, implementation, monitoring, reformulation and evaluation. Examples will be provided on how the livelihood framework operates in different contexts, such as in emergency, mitigation, recovery and development settings. Emphasis will be placed on how participatory approaches have been integral to this process. In addition to the use of the livelihood framework in new programme designs, the paper will document experiences dealing with retrofitting existing projects/programmes that originally did not use a livelihood framework. Lessons learned from this paper will provide future guidance.
During the past several years, much conceptual progress has been made in an understanding of the processes that lead to household food insecurity (Frankenberger 1992). In the 1970s, food security was linked mostly to national and global food supplies. The food crisis in Africa in the early 1970s stimulated a major concern on the part of the international donor community regarding supply shortfalls created by production failures due to drought and desert encroachment (Davies et al. 1991). This focus on food supplies as the primary cause of food insecurity was given credence at the 1974 World Food Conference.
A focus on household food security with an emphasis on food access (1980s). The limitations of the food supply focus came to light during the food crisis that plagued Africa in the mid-1980s. It became clear that adequate food availability at the national level did not automatically translate into food security at the individual and household levels. Researchers and development practitioners realized that food insecurity occurred in situations where food was available but not accessible because of an erosion in people's entitlement to that food (Borton and Shoham 1991). Sen's (1981) theory on food entitlement had a considerable influence on this change in thinking, representing a paradigm shift in the way that famines were conceptualized. Food entitlements of households derive from their own production, income, gathering of wild foods, community support (claims), assets, migration, etc. Thus a number of socio-economic variables have an influence on a household's access to food. In addition, growing food insecurity was viewed as an evolving process where the victims were not passive to its effects. Social anthropologists observed that vulnerable populations exhibited a sequence of responses to economic stress, giving recognition to the importance of behavioural responses and coping mechanisms in food crises (Frankenberger 1992). By the late 1980s, donor organizations, local governments and NGOs began to incorporate socio-economic information in their diagnoses of food insecurity.
The household food security approach that evolved in the late 1980s emphasized both availability and stable access to food. Thus, food availability at the national and regional level and stable and sustainable access at the local level were both considered essential to household food security. Interest was centred on understanding food systems, production systems, and other factors that influenced the composition of food supply and a household's access to that supply over time. What was not clear was how nutritional outcomes were factored into food-security deliberations.
A focus on nutritional security with an emphasis on food, health and mother and child care (early 1990s). Work on the causes of malnutrition demonstrated that food was only one factor in the malnutrition equation, and that in addition to dietary intake and diversity, health and disease, and maternal and child care were also important determinants (UNICEF 1990). Household food security is a necessary but not sufficient condition for nutritional security. Researchers found that there were two main processes that had a bearing on nutritional security. The first determined access to resources of food for different households. This was the path from production or income to food. The second process involved the extent to which the food obtained was subsequently translated into satisfactory nutritional levels (World Bank 1989). A host of health, environmental, and cultural/behavioural factors determine the nutritional benefits of the food consumed; this is the path from food to nutrition (IFAD 1993).
This work on nutritional security demonstrated that growth faltering could not necessarily be directly related to a failure in household food security. It shifted the emphasis away from simple assumptions concerned with household access to food, resource base and food systems by demonstrating the influence of health and disease, "caring" capacity, environmental sanitation and the quality and composition of dietary intake on nutritional outcomes.
A focus on household livelihood security (1990s). Research carried out in the late 1980s and early 1990s indicated that the focus on food and nutritional security as they were currently conceived needed to be broadened. It was found that food security was but one subset of objectives for poor households, and only one of a whole range of factors that determined how the poor made decisions and spread risk and how they finely balanced competing interests in order to subsist in the short and longer term (Maxwell & Smith 1992). People may choose to go hungry to preserve their assets and future livelihoods. Therefore, it is misleading to treat food security as a fundamental need, independent of wider livelihood considerations.
Thus, the evolution of the concepts and issues related to household food and nutritional security led to the development of the concept of household livelihood security. The HLS model adopted by CARE allows for a broader and more comprehensive understanding of the relationships among the political economy of poverty, malnutrition, and the dynamic and complex strategies that the poor use to negotiate survival. The model places particular emphasis on household actions, perceptions and choices, with food understood to be only one of the many priorities. People are constantly being required to balance food procurement against the satisfaction of other basic material and non-material needs (Maxwell & Frankenberger 1992).
To summarize, there were three strategic shifts in development thinking that led CARE to the adoption of a livelihood approach:
Many of the definitions of livelihood security currently in use derive from the work of Chambers & Conway (1992). A livelihood "comprises the capabilities, assets (stores, resources, claims, and access) and activities required for a means of living; a livelihood is sustainable which can cope with and recover from stress and shocks, maintain or enhance its capabilities and assets, and provide sustainable livelihood opportunities for the next generation" (Chambers & Conway 1992).
Household livelihood security has been defined as adequate and sustainable access to income and resources to meet basic needs (including adequate access to food, potable water, health facilities, educational opportunities, housing and time for community participation and social integration) (Frankenberger 1996). An attempt is now being made in CARE's work to shift to more rights-based approaches, which place more emphasis on access issues and the policy environment and which treat people more as active beings1 (see Figure 1). Livelihoods can be made up of a range of on-farm and off-farm activities that together provide a variety of procurement strategies for food and cash. Thus, each household can have several possible sources of entitlement, which constitute its livelihood. These entitlements are based on the endowments that a household has and its position in the legal, political and social fabric of society (Drinkwater & McEwan 1992). The risk of livelihood failure determines the level of vulnerability of a household to income, food, health and nutritional insecurity. The greater the share of resources devoted to food and health service acquisition, the higher the vulnerability of a household to food and nutritional insecurity. Therefore, livelihoods are secure when households have secure ownership of or access to resources (both tangible and intangible) and income-earning activities, including reserves and assets, to offset risks, ease shocks and meet contingencies (Chambers 1988). Households have secure livelihoods when they are able to acquire, protect, develop, utilize, exchange and benefit from assets and resources (Ghanim 2000).
Source: After Swift, 1989; Drinkwater, 1994; Carney, 1998; Frankenberger and Drinkwater, 199
The idea of household livelihood security as defined above embodies three fundamental attributes: (1) the possession of human capabilities (e.g. education, skills, health, psychological orientation); (2) access to other tangible and intangible assets (social, natural, and economic capital); and (3) the existence of economic activities (Drinkwater & Rusinow 1999). The interaction among these attributes defines what livelihood strategy a household pursues and is thus central to CARE's household livelihood security model.
CARE fully recognizes that its partners (donors, host government counterparts and local civil-society- and community-based groups) may be using or adhering to different development approaches (models or frameworks). While its language and content may differ to varying degrees, CARE believes that the HLS framework is compatible with other development approaches because of the principles that underscore it (Beckwith 2000). These include:
In its simplest form, livelihood security is the ability of a household to meet its basic needs (or realize its basic rights). These needs include adequate food, health, shelter, minimal levels of income, basic education and community participation. If any of these basic needs is not met, CARE considers that household to be living in absolute poverty (Frankenberger 1996). However, simply satisfying people's basic needs is not adequate to ensure that those people can rise above and stay above absolute poverty (Beckwith 2000).
For CARE, sustaining livelihood security depends on a number of enabling conditions being in place. These include human rights recognition, civil participation/action, risk management, an enabling policy environment, gender equity and environmental stewardship. By contributing to the establishment of this enabling environment, CARE hopes to assist people in meeting their basic needs on a sustained basis. CARE believes that these elements are the underpinnings of its vision, " ... a world of hope, tolerance and social justice, where poverty has been overcome and people live in dignity and security".
CARE has faced a number of challenges in trying to institutionalize a livelihood approach. This transition has resulted in enormous debate and feedback from the field as country offices have tried to operationalize the concept.
The household livelihood security concept was first introduced in 1994 by the Food Security Unit at CARE's headquarters. Using design opportunities presented by the development of new United States Agency for International Development (USAID) Title II-funded programmes and other donor resources, an attempt was made to use a holistic diagnostic approach in the designing of livelihood security programmes. Since then, multisectoral teams have conducted rapid or participatory livelihood security assessments in countries as many and as diverse as Afghanistan, Angola, Azerbaijan, Bangladesh, Benin, Bolivia, El Salvador, Ethiopia, Georgia, Guatemala, Haiti, Honduras, India, Madagascar, Malawi, Mali, Mozambique, Nepal, Nicaragua, Niger, Rwanda, Somalia, Sri Lanka, Sudan, Tajikistan, Tanzania, Togo, Zambia, and Zimbabwe.
After the first wave of assessments was carried out in 1995-1996, many country offices became interested in the approach, although the assessment methodology at that stage was viewed as being too expensive and extractive. Tensions were also created when some country offices felt that the framework was being imposed on them from headquarters. Confusion existed as to the objectives of these assessments, and whether the HLS approach was simply an assessment methodology or a whole project process framework.
Numerous reflective discussions and workshops were held in each of the regions that focused on the lessons learned from the application of HLS diagnostic tools. This enabled each region and country office to adopt its own context-relevant approach for implementing a livelihood security framework. Many countries opted for smaller assessments that were more participatory and less quantitative. The challenge has been to allow for this flexibility and creativity and at the same time ensure that country offices are adhering to bottom-line principles (these are discussed later).
Similarly, at headquarters many of the sector specialists also felt that the livelihood framework was being imposed on them by senior management. This created some resistance to its adoption. Much of this resistance could have been avoided through a more inclusive process in the beginning.
In some countries, the donors were not very receptive to holistic design processes, particularly if they had had a sector bias in funding. For example, CARE India and CARE Bolivia had some difficulty convincing their major donor that holistic assessments and multisectoral designs were appropriate. Changes in staff and programming direction within the donor organization allowed for the more holistic programming to be brought in later. In Nepal, holistic programming was achieved by having different donors fund different sectoral activities targeted to the same remote areas. There is still some concern among some donors as to the compatibility of sector-wide approaches and livelihood approaches.
Sector biases of government ministries with whom CARE is aligned also can present difficulties in cross-sectoral programming. For example, in India, CARE works with the ministry that oversees the Integrated Child Development Services (ICDS) safety net programme. This ministry does not deal with agricultural development. If CARE wants to expand its activities into improving agricultural productivity of the poor, it will have to align itself with a different part of the Indian Government. Such agreements are not always easy to establish.
In the beginning, many of the interventions that were implemented were along traditional lines. More recently, CARE country offices have tried to take into consideration other cross-cutting social and political issues that have been hindering the poor from achieving livelihood security. Policy advocacy and rights-based approaches are now starting to be integrated into programming activities. Such programming changes demonstrate CARE's recognition that poverty is not only a matter of inadequate access to income, food and services, but fundamentally a social and political issue too.
One key question that continues to be asked by some CARE staff is: What are we really gaining through the application of a livelihood security approach? The responses that have come back from the field with regard to the value added by this approach include that HLS:
At all levels of programmatic decision-making, the HLS framework is CARE's point of departure (see Figure 2). The various lenses embedded within the livelihood framework assist CARE in its analysis of a given situation or geographical area, whether it is working at the strategic, regional, programme, project or sector level.
Source: Modified from Scoones, 1998
Livelihood profiles are derived for a country or region through analytical lenses that are clustered under the following categories: contexts, conditions and trends; livelihood resources (economic, natural, human and social capital); institutional processes and organizational structures (government, civil society and private sector); livelihood strategies (productive and exchange activities); and livelihood outcomes (e.g. nutritional security, food security, health security, habitat security, education security, income security, social network security, safety and environmental security).
Context, conditions and trends. A holistic analysis of livelihood security begins with understanding the context for any given population. To understand the macro-level factors that influence the range of possibilities for livelihood systems, we must consider the social, economic, political, environmental, demographic, historical, and infrastructural information. It is this information that sets the parameters within which livelihood strategies operate. To reduce costs, this information is primarily derived from secondary data.
Livelihood resources. Households have access both to tangible and to intangible assets that allow them to meet their needs. Natural capital consists of natural resource stocks from which resource flows that are useful for livelihoods (e.g. land, water, wildlife, biodiversity and environmental resources) are derived. Social capital is the quantity and quality of social resources (e.g. networks, membership in groups, social relations and access to wider institutions in society) upon which people draw in their pursuit of livelihoods and as safety net mechanisms for meeting shortfalls in consumption needs. The quality of the networks is determined by the level of trust and shared norms that exist among network members. People use these networks to reduce risks, access services, protect themselves from deprivation and acquire information to lower transaction costs. Human capital consists of skills, knowledge, good health and the ability to labour, which are important for the pursuit of livelihood strategies. Economic capital is the productive resources and stores (e.g. savings, credit, remittances, pensions), basic infrastructure (e.g. transport, shelter, energy, communications and water systems), production equipment and other means that enable people to pursue their livelihoods.
In an analysis of these resources, it is important to take into account the combinations necessary for sustainable livelihoods, the trade-offs that exist between resources, the sequences that may exist between them (i.e. which resources are prerequisite for others) and the long-term trends in their use (adapted from Scoones 1998).
Institutional process and organizational structures. A number of institutions operate in the community milieu that influence livelihood outcomes. The State not only provides services but also offers safety nets, changes policies and limits freedoms that can have positive or adverse effects on livelihood systems. Similarly, formal civil-society organizations (NGOs, CBOs, parastatals, cooperatives, churches) can provide enabling conditions or constrain opportunities for households. Informal civil society (e.g. informal community networks) consists of the web of networks to which individuals and households belong. These networks can have positive or negative influences on the livelihood strategies that people pursue. The private sector can also create or limit households' opportunities. In the formulation of any sustainable interventions it is important to take these various institutions into account.
Livelihood security strategies. Households combine their livelihood resources within the limits of their context and use their institutional connections to pursue a number of different livelihood strategies. Such strategies can include various types of production and income-generating activities (e.g. agricultural production, off-farm employment, formal sector employment) or a combination of the two. An HLS analysis should determine the livelihood strategy portfolios that different households pursue and the historical pathways they have taken.
Livelihood security outcomes. To determine whether or not households are successful in pursuing their livelihood strategies, it is important to look at a number of outcome measures that capture need or well-being satisfaction. Nutritional status is often considered one of the best outcome indicators for overall livelihood security since it captures multiple dimensions, such as access to food, health care and education. Other livelihood outcomes that should be measured include sustained access to food, education, health, habitat, social network participation, physical safety, environmental protection and life skills capacities. Analysis of these outcomes should determine not only what needs are currently not being met but also what trade-offs there are between needs. In addition, such an analysis should help determine the synergistic relationships among these outcome measures.
In addition to these standardized measures, attempts are made to derive from the community the criteria they use for determining livelihood improvement. These measures are often location specific. Every effort is made to establish community-based monitoring systems to enable the community to track improvements themselves.
CARE is currently trying to establish these livelihood profiles during the long-range strategic planning process for each region in which it operates (e.g. Bolivia) or in analytical work conducted as part of a programme design exercise. This ensures that a more holistic perspective is taken in any project design for that region, even if a short time horizon is provided for the development of a particular proposal for a donor. This will allow CARE to take a more holistic perspective in any project design for a given region, even if it is given an extremely short time horizon to develop a proposal for a donor. These profiles would be periodically updated as new information comes in from projects. The framework provides a way to organize that information.
A further area of increased exploration within the context of CARE's work in recent years is the generation of a growing range and intensity of operational relationships with other organizations. Gone are the days when CARE saw itself primarily as an organization responsible for the direct delivery of goods and services to those affected by emergencies, and to the poor and vulnerable in general. There are multiple reasons for this, but some of the pre-eminent ones are:
These factors in particular have caused CARE to see its international responsibility increasingly in terms of seeking to influence and, in return, learn from and collaborate with a growing number of agencies of different types and hues. This role is being played out at all the levels at which CARE operates: internationally, regionally, nationally and more locally within country contexts. New programme approaches with an increased emphasis on partnership and multi-agency collaboration are being generated, with CARE's role often being to facilitate the creation of linkages between community-based and other CSO actors, government and private-sector agencies who have not worked together previously.
From CARE's perspective, partnerships are defined as "mutually beneficial alliances of diverse types between organizations where roles, responsibilities and accountabilities are clearly defined. Partnerships facilitate continuous two-way learning and are based on trust, shared vision and commitment to common objects. Partnership is a means to achieve improved quality of life for more beneficiaries through sustainable service delivery, better responsiveness to local development needs, and increased scale and scope of programmes".
In terms of vision, "CARE will strive to be a reliable and trusted partner with an enhanced reputation and ability to improve the livelihood security of poor households through a diverse dynamic network of partners. In every intervention, CARE will explore linkages that reach greater numbers of people, alleviate poverty and save more lives" (Beckwith 2000).
The major objectives for CARE's partnering are to:
The major partnership principles advocated by CARE are to:
One example of such partnership relationships is the Strengthening Capacities for Transforming Relationships and Exercising Rights (SCAPE) project in South Africa. South Africa is a country of multiple institutions that often have limited capacity and limited scope to their activities. This applies to many parts of the country's complex and cumbersome three-tiered government structure, as well as to civil society. Both are struggling to adapt to the changes wrought by the coming of a democratic government and society in 1994, which has resulted in a process whereby the country's non-white population has gained rights it previously lacked, but where old attitudes and practices hinder the evolution of more empowering development approaches. This has created a situation that does not enable people to gain the confidence and understanding of how to exercise their new rights so as to benefit their livelihoods. This applies equally to local communities, to civil-society organizations working with them and to local government, all of which retain an expectation that resources and solutions will be provided centrally. Accordingly, over the space of two years, the CARE South Africa office has been developing and piloting a programme in which it works with multiple partners in transforming the horizontal and vertical relationships that affect the nature and effectiveness of local development policies.
More commonly in CARE now, many country offices are working in partnership with municipal governments. For example, in Latin America, both CARE Bolivia and CARE Honduras have been working with municipal governments in their project areas, focusing on strengthening planning and service delivery. A recent evaluation of the programme in Bolivia found that municipal partners were extremely effective institutions for promoting HLS programming. This is because these institutions are holistic in their service delivery. Similarly, in southern Africa, urban livelihood programmes have established successful partnerships with municipal authorities in Angola, Madagascar, Mozambique and Zambia.
Much of CARE's partnership efforts involve working with civil-society groups. This is illustrated by the agency's work with local NGOs in Somalia and South Africa and with community-based organizations in Mali and Zambia. In its most tangible form, civil society is defined by CARE as the range of institutions and organizations that represent individual citizens or that provide people the means by which to connect themselves collectively to government or the private sector. Civic action is the dynamic and collaborative relationship among citizens, government and the private sector that contributes to the well-being of individual citizens. For CARE, a strong civil society means the ensuring of a dynamic and beneficial relationship between the institutions and organizations that represent government, the private sector and civic groups. CARE's civil society strengthening efforts include: (1) building organizational capacity and strengthening institutions; (2) supporting mechanisms for dialogue and advocacy among the three sectors of society; (3) increasing the effectiveness and synergy among these institutions for the benefit of individual citizens; and (4) promoting the inclusion of the poor, disenfranchised and marginalized citizens in the enjoyment of benefits derived from civic action (Beckwith 2000).
Strengthening civic action to promote household livelihood security involves strengthening government, the private sector and civic groups in order to help the poor reduce risk, improve access to services and lower transaction costs. Institutional analyses carried out in programme design should help determine the weak institutions that need to be strengthened. To be effective, each sector must be able to manage risk as well as perform the complementary functions it is supposed to perform during non-crisis years. Risk management is one aspect that has not been taken into account in most CARE institutional assessments.
Long-range strategic planning (LRSP) exercises have been carried out for every country in which CARE works. These exercises are normally developed for a five-year period, unless a given country is under emergency conditions, which entails shorter planning horizons (often of two years). The HLS framework has been used in this planning process for organizing data on vulnerable groups in different geographical areas; causal explanations regarding shocks, trends and processes; macro-micro linkages that are key to understanding the programming areas; and institutions with which CARE will create alliances within programme implementations. The trend over the last couple of years has been to move from descriptive and impressionistic summaries to analytical processes and syntheses of priorities. The LRSP helps the country office align its programming where the need is great, where the potential for partnering with local institutions is high and where CARE has a comparative advantage. Secondary data are primarily used in this planning process.
Within the Latin America region, CARE Guatemala used the HLS framework in crafting its 1998-2002 Long Range Strategic Plan. Similarly, Haiti, Honduras and El Salvador modified their information-gathering and analysis using an HLS framework. As part of these planning processes, it was perceived by each of the country offices that certain structural adjustments would be required to implement HLS programming. Such structural changes included the creation of regional decentralized structures that would allow for multisectoral programming within a given geographical area.
After reviewing the lessons learned from these structural changes, country offices in Latin America are now realizing that structural changes do not necessarily lead to better HLS programming. HLS programming does not always favour one type of structure over another. Centralized sector-based structures and regional structures both can promote coherent HLS programming. What is important is to use the HLS framework to target CARE's interventions more effectively in order to achieve leverage, synergy and cost efficiencies. A variety of team management styles can be put together to achieve these objectives. One of the key bottlenecks facing most country offices in implementing HLS programming is their having sufficient technical expertise both to service specific geographical area demands and to establish a significant level of consistency across geographical areas regarding programming approaches and methodologies (Beckwith 1999).
Another area of concern expressed by country offices is that they are tending to grow in geographical scope and complexity vis-à-vis multisectoral programming wherever they operate. Application of the HLS framework should help them clarify their rationale for working in specific geographical areas and, hence, consolidate their portfolio and promote more focused targeting of interventions (Beckwith 1999). When programmatic decisions are not focused through the use of the framework, a less strategic growth of programmes can put considerable strain on staff and management.
The need for holistic analysis as a basis for a livelihood approach often engenders nervousness in programme staff, who fear that it implies a lengthy, in-depth and complex process (Drinkwater & Rusinow 1999). Alternatively, the livelihood analysis might take on a life of its own; indeed it may become an end in itself. Both of these dangers can and should be avoided, as it is critical to minimize gaps between the analysis and design stage as well as avoid unnecessary data collection and maintain an interactive relationship with stakeholders (Drinkwater & Rusinow 1999).
A wide range of tools can be used - from a quick situational analysis to an in-depth or geographical-wide analysis of livelihoods - to determine the causes of vulnerability and the extent of poverty. The key is to ensure that emphasis is placed on gaining a multidimensional view of livelihoods that allows for the identification of the most vulnerable households, and on placing people's priorities and aspirations for improving their livelihoods firmly at the centre of the analytical and planning process (Drinkwater & Rusinow 1999).
Over the past five years, rapid or participatory livelihood security assessments (RLSAs or PLSAs) have become major tools for the collection and analysis of this type of information and, therefore, a major means of operationalizing an HLS approach. The main purpose of these participatory assessments is to understand the nature of livelihood strategies of different categories of households (social differentiation), their levels of livelihood security and the principal constraints and opportunities to address through programming. This information is also disaggregated by gender and generation. Therefore, a good holistic analysis would develop an understanding of livelihoods that was contextual, differentiated and disaggregated. The methods used often focus on visualizing information, with community members involved as much as possible in documenting information. Outputs from such assessments at a minimum include the identification of the risk factors facing households, the key location-specific criteria for differentiating wealth categories of households and the key leverage points and opportunities to pursue in future programming.
In terms of the distinction between methodologies, with a maximum of two days spent per site or area, rapid assessments usually achieve a broader scan over a wider area where little secondary data exists and where a major shock or rapid change has occurred. Participatory assessments offer a more in-depth analysis of fewer communities and are usually undertaken where some of decisions about the likely geographical location of any ensuing programme activities have already been made.
Objectives and information requirements. The most common objective of livelihood security assessments (LSAs) is to acquire information for the design of programmes. However, most LSAs have multiple objectives. One objective may be global learning to gain institutional credibility in an area where there is little or no previous experience, or to get information for strategic planning to improve the allocation of scarce programme resources over multiple, competing demands. Building the analytical capacity of staff and partner organizations is often an objective, though rarely a primary one. Building partnership relationships is also a common secondary objective (Figure 3). An important consideration in setting objectives is whether programmes based on information gathered will be scaled up within the planning time horizon. How much primary information must be collected depends on the availability and quality of existing information. In general, the principle is to collect only as much primary information as is required that cannot be gathered from secondary sources (Figure 3).
The analytical framework generally defines the types of information required and includes qualitative descriptive information, quantitative descriptive information and analytical (or causal) information (see Table 1). The use of this framework has recently been applied in rural assessments in Malawi and Zimbabwe, and in urban assessments in Mozambique and Peru.
Qualitative descriptive information: At the household level, the information primarily required includes the assets held by the household and how these are used to earn adequate income, how resources are allocated and the levels of critical outcomes achieved in terms of food security, nutrition and health status and access to other basic needs such as water, shelter and education. Assets include not only productive assets, such as land and livestock, or financial assets, such as savings or cash, but also the more intangible assets of labour, skills, capacity and the social relations that underpin livelihood activities. Important among these is the ability of some households to cope better than others with risk and crisis, what these abilities are, and how coping strategies work. At the intrahousehold level, it is important to consider gender and generationally differentiated roles and responsibilities, power relations and differential access to resources and opportunities. Livelihood systems must be understood at the community as well as the household level. Household-level outcomes have to be put in a community or broader social and political context, so general information on the social, political, and institutional environment is also a major requirement.
Quantitative descriptive information: For geographic targeting, and for identifying vulnerable groups, quantitative indicators of household basic needs outcomes are required. These will include nutritional status information as well as information on health status, access to services, literacy levels, access to potable water, etc. Much of this information is obtained from secondary sources.
Analytical (causal) information: For effective programme design, it is important to understand not only the current status of target groups but also the sources of vulnerability and the causal factors that lead to vulnerability.
Household livelihood security analytical framework for programme design, implementation and evaluation
Design & implementation
Understanding vulnerability risk factors
To understand vulnerability, it is important to take into account the shocks or risks to which households are exposed, their ability to cope with those shocks and their resilience to future shocks. To determine this vulnerability, risk factors can be grouped into those that are:
Sources of risk to household livelihood security
Sources of livelihood
Labour power, education, health
Disease epidemics (malaria, cholera, dysentery) due to poor sanitary conditions, AIDS
Declining public health expenditures, user charges, declining education expenditures
Breakdown in community support of social services
Privatization of social services, reduction in labour opportunities
Conflict destroys social infrastructure, mobility restrictions
Financial and natural capital
Productive resources (land, machinery, tools, animals, housing, trees, wells, etc.), liquid capital resources (jewellery, granaries, small animals, savings)
Drought, flooding, land degradation, pests, animal disease
Land confiscation, insecure tenure rights, taxes, employment policies
Appropriation and loss of common property resources, increased theft
Price shocks, rapid inflation, food shortages
Conflict leads to loss of land, assets, and theft
Claims, kinship networks, safety nets, common property
Recurring environmental shocks breakdown ability to reciprocate; morbidity and mortality affect social capital
Reduction in safety net support (school feeding, supplementary feeding, FFW, etc.)
Breakdown of labour reciprocity, breakdown of sharing mechanisms, stricter loan requirements, lack of social cohesion
Shift to institutional forms of trust, stricter loan collateral requirements, migration for employment
Communities displaced by war; theft leads to breakdown in trust
Sources of income
Productive activities, process and exchange activities, other sources of employment, seasonal migration
Seasonal climatic fluctuations affecting employment opportunities, drought, flooding, pests, animal disease, morbidity and mortality of income earners
Employment policies, declining subsidies or inputs, poor investment in infrastructure, taxes
Unemployment, falling real wages, price shocks
Marketing channels disrupted by war
Once the risks have been taken into account, it is important to understand how households cope with or adapt to these shocks. On the basis of this analysis, it is possible to determine trends, livelihood strategies and changes that occur in internal household dynamics. It is important to determine also the role of social networks and institutions in adapting to and coping with these changes and to analyse the intra- and intercommunity dynamics.
On this basis, we can determine vulnerability at the community, household and individual level. This analysis delineates the target populations that need to be focused on in future interventions.
In addition to capturing unintended effects after the fact through participatory monitoring and evaluation, CARE has begun to put much more emphasis on predicting - and mitigating - unintended effects through better programme design. Growing out of the desire, particularly in emergencies, to "do no harm", CARE has developed over the past year a set of tools intended to enable it to conduct a benefits/harms analysis prior to beginning an intervention.2 Similarly, the organization is also improving its gender analysis methodologies.
The HLS approach permits CARE to plan for and build on positive cross-sectoral impacts. The benefits/harms tools are intended to predict, minimize and mitigate cross-sectoral negative impacts. Under an emergency circumstance, for example, a food project may create dependency and undermine self-reliance, or it may make people targets for raiding. The benefits/harms analysis permits consideration of the overall impact of interventions - both within the sector of focus and beyond.
The framework organizes cross-sectoral impacts into five categories, and analyzes the different reasons for cross-sectoral impacts. The five categories include:
CARE's Benefits-harms handbook, developed in East Africa, offers three different types of tools:
Over the past five years, about 30 major LSAs, as well as a number of more limited exercises (where, for example, a secondary review and limited field information collection sufficed to fulfil the objectives of the exercise), have been carried out by CARE worldwide. A few of the salient lessons learned include:
Perhaps the largest challenge CARE has faced institutionally in operationalizing a livelihood security approach is developing a framework that is at once inclusive enough to facilitate natural variation in its application, depending on both the context and the programming instincts of those involved, and at the same time provides definite guidance on what are perceived as bottom-line principles.
CARE has evolved a key set of elements or principles of programme quality. These principles are illustrated in Figure 4. Several variations of this diagram have been produced over the last year, although the key elements are similar. That is, within CARE's design framework, a programme should contain a holistic analysis, a synthesis, a focused strategy, a coherent information system and reflective practice.
There are three important points to be noted about this design framework. The first is that it is a framework intended to improve the quality of CARE's programmes. Thus for instance, within the southern and West African regions, this diagram provides the central conceptual focus of current efforts across the region to improve programming. The variation of the diagram illustrated here was produced as the organizing focus of a regional design workshop held in January 2000, and is also being used as an organizing focus for the programme sections of the region's annual operating plan (AOP). This means that there is an increasing focus on programme development work in country offices in the region being organized around this framework.
The second point to be noted about this design framework is that although CARE is operationalizing it through the use of a livelihood security framework, this is not an inherent requirement of the framework. The importance of this is that other conceptual approaches can accordingly also be deployed in the practical use of the framework. This is partly shown by the reference to "participation", "partnerships" and "personal and social empowerment" in the second circle of the diagram. Emphases on all these aspects are features of most programmes in the region using a livelihoods framework, but these also bring their own conceptual and methodological tools into the design process, which often provide the "vehicle" within which the livelihood framework is deployed. Similarly, at an advocacy workshop organized in Sussex in October 1999, others in CARE showed how different analytical lenses - livelihoods, human rights, stakeholder, policy analysis - could all be used in the context of the framework. For the livelihoods framework to be seen as inclusive in this way, enabling a variety of approaches to be evolved and utilized, it is critical to promoting a broad institutional practice of the framework.
This situation allows for the development of healthy organizational debate over the nature of the methodologies being used and what constitutes good practice in terms of conducting holistic analysis and synthesis, developing a focused strategy and coherent information system and being reflective in practice. The points noted under each element of the framework illustrate the developing practice. For example, it has already been stated that if an understanding of livelihoods is to be developed during a holistic analysis, there must be an analysis of context (how the household and community relate to the wider world), of the differentiated nature of livelihoods (livelihood categories) and of the disaggregated situations of different individuals within the household (gender and generational roles and issues). In addition, other analytical lenses are also commonly deployed: an understanding of different stakeholder perspectives is developed, an institutional assessment conducted and human rights issues explored either in conjunction with or separately from an examination of basic needs, and so on.
The third point to note about the framework is that it is both iterative and non-linear. There is a "to-ing" and "fro-ing" between the different elements of the framework, both before and after the project is formally approved. For instance, a participatory livelihoods assessment exercise may be carried out, as part of the analysis stage, but may conclude with methods that lead into synthesis and strategy design. Similarly, some methods may also provide provisional ideas on livelihood indicators, which are then developed more fully during the strategy design and when the information system is being developed more fully during implementation.
An attempt to illustrate this is contained in Figure 5. The diagram also aims to show that the point at which the project is formally funded may fall at different times in the process, depending on the status of the factors in the "variables" list. Thus, for example, Mahavita, an urban livelihoods project in Antananarivo, Madagascar, was funded for five years after a relatively brief secondary review and participatory livelihood assessment exercise. As a result, a great deal more detailed analytical work was also conducted during the project's start-up phase. In contrast, other projects may require much more protracted negotiation work, and the process to proceed as far as the piloting of activities, before there is any guarantee of more secure funding.
One of the key distinctions between a livelihoods approach and, for example, the integrated rural development programmes of a decade ago, is that in the livelihoods approach the holistic analysis should give rise to a focused strategy rather than to a broad range of inadequately linked activities. The synthesis stage of the analytical activities should be used to build hypotheses on what are likely to be the three or four major project components of "lines of action" that will have the greatest leverage or beneficial impact on improving livelihoods. Koos Neefjies of Oxfam labelled this the acupuncture approach, because "a good acupuncturist uses a holistic diagnosis of the patient followed by very specific treatment at key points. Holistic diagnosis does not mean needles everywhere!" (Ashley & Carney 1999, 17).
The types of projects developed at CARE using a livelihoods framework are diverse. Some are of a more multisectoral, or disciplinary, nature, but applying a livelihoods approach does not preclude projects being largely of a sectoral nature. What is important is that a holistic perspective is used in the design to ensure that cross-sectoral linkages are taken into account, and that the needs addressed in project activities are those that deal with the priority concerns of households and that build upon the experience and coping mechanisms they have evolved (Drinkwater & Rusinow 1999). Increasingly among some donors there is also expected an acceptable cost-benefit ratio of resource use, which again emphasizes the issue of key leverage points being identified, which can be expected to lead to the greatest beneficial impact on livelihoods, whatever the type of programme.
CARE also has substantial experience in using a livelihoods framework to redesign or reorient existing project activities. There are several examples of this in the southern and West African regions. In all cases, the reorientation process requires a return to the analytical basis for the project or programme and reworking this using the livelihoods framework. In practice this usually means conducting some form of livelihood assessment, coupled with a reconsideration of secondary information and the contextual analysis. One example is the Training for Agricultural and Environmental Management (TEAM) project in Lesotho. Following a two-year pilot, which had not used an HLS approach, a redesign of the programme was carried out that was based largely on a series of village-level participatory livelihood assessments. All told, 46 of these were conducted, of which 3 contributed directly towards the redesign. The remainder were conducted after the funding of the new phase, in part with the intention that they form baseline exercises in the village. In addition, the new two-year phase had an action research component that provided a more in-depth understanding of livelihoods in three distinct village areas, and that also contributed to the design of the project information system. This action research component aided considerably the further modification of the programme at the conclusion of the second two-year pilot.
A second and more thorough illustration of the use of a livelihoods framework in redesign comes from Zimbabwe. In a process that is still ongoing, detailed participatory livelihood assessments were conducted in November 1999 at four sites in the Midlands and Masvingo Provinces. Three of these were areas in which CARE Zimbabwe was already operating, therefore the assessment methodology included visioning activities at both the community and the team level. At the community level, the aim was to generate ideas on what people saw as the main opportunities and priorities for improving their livelihood security. The team members were then able to use this information, together with their knowledge of the nature of the programme in each province, to produce ideas on how the programme might be more effective in its impact on livelihood security. The process produced consensus at the community and team level on a clear set of ideas, and the issues that would be inherent in their realization, for progress.
The assessment information is now being incorporated within a strategic follow-up process that seeks to improve the nature and effectiveness of existing project activities, develop synergies across these, build adherence to a common set of programming principles, provide a coherent basis for the development of further activities and, from the programme restructuring, lead to a complementary administrative restructuring of the sub-offices.
One area that CARE has been seeking to incorporate more successfully into its livelihood framework is that of gender. Experience from some of CARE's projects in Zambia, which have been using a livelihoods framework since 1995, has shown that working towards achieving greater gender equity in the benefits of programmes is an especially difficult challenge. Dealing with the issue at all requires a project to have both a well-developed information system and the ability to engage in internal reflection and learning. In fact, the "reflective practice" element of the programme design framework was added during a workshop in which CARE Zambia staff discussed their experiences using HLS and gender frameworks, examining what the major programming issues were with regard to both, and how they could move forward more effectively with a better integrated "HLS + gender" approach.
Gender equity in CARE programming entails the condition of fairness in relations between men and women, leading to a situation in which each has equal status, rights, levels of responsibility and access to power and resources. Gender is considered different from sex, which describes the universal biological characteristics of men and women. Gender refers to the socialization process that assigns certain attitudes, roles and responsibilities to men and women, and results in different opportunities and behaviour for each. It is dynamic, varying within and among societies, and over time, and is influenced by cultural, economic, political and environmental factors. CARE seeks to ensure that change brought about by programmes responds to mutually agreeable standards of fairness of both women and men in their given contexts (Beckwith 2000).
Different from sameness, equity is based on the concept of what is just, and the premise that women and men, by virtue of their common humanity, deserve equal opportunities to define their paths in life. It does not prescribe a division of roles, nor does it ignore the fact that success will ultimately rest with the inspiration and efforts of each individual. CARE's focus on equity is a recognition that in much of the world, opportunities are not equal and the playing field is not level (Beckwith 2000).
Household livelihood security continues to be the cornerstone framework that CARE uses to carry out its programming efforts. It is considered an organizing framework that is used systematically to inform decisions and as a desired end state or goal for CARE's programmes. The HLS framework enables a more holistic view of the world to inform CARE's programming decisions. The root causes of poverty can be understood more clearly, as can the opportunities and leverage points for positive change. Application of the livelihood framework should not be considered a linear process, but rather a flexible, dynamic and iterative process over time.
While HLS is predicated on taking a holistic view of a given situation, different entry points can be used for its application. In addition, taking a holistic view does not always mean undertaking multiple interventions. Ongoing sector-focused interventions may be modified or interrelated to incorporate a livelihood security perspective. Alternatively, a single-sector intervention may provide the key leverage activity in a given situation.
Over the past two years, CARE has identified several analytical lenses that have been incorporated into an HLS holistic analysis in order for the agency to understand better the root causes of poverty. These analytical lenses include basic needs, a human rights perspective, civil participation and action, gender and the policy environment. Numerous staff within CARE are now working on the tools and methods that will allow the country offices to incorporate these analyses in programme decision-making.
The broader and more in-depth understanding brought about by the application of these analytical lenses will expand CARE's programming choices regarding what is done, how it is done, who it is done with and who benefits. These analytical tools will contribute significantly to the future directions of CARE programming
In the end, the HLS framework is helping CARE make strategically focused choices about where to concentrate its limited resources and its comparative advantages in order to leverage the most positive and lasting change. It is through these efforts that CARE will continue to contribute to the global effort to end poverty.
1 See for example Clare Ferguson, 1999, "Global social policy principles: human rights and social justice", on the active/passive difference between rights-based and needs-based approaches. Personal communication. CARE Bolivia.
2 P. O'Brien. 1999. Benefits-harms handbook, Nairobi. CARE East Africa.
3 D. Maxwell, 1999, "Livelihoods and vulnerability: how different is the urban case?" Presented to 1999 Meeting of the Society for Applied Anthropology. Tucson, Ariz.