Enable low bandwidth mode Disable low bandwidth mode

Introduction (Special Supplement 1)



ACF Action Contre la Faim
BMI Body Mass Index
CMT Community Managed Targeting
CSB Corn Soya Blend
DFID Department for International Development
FFW Food for work
HEA Household Economy Approach
HPN Humanitarian Practice Network
MT Metric tonne
MSF Médecins Sans Frontières
MUAC Mid upper arm circumference
NGO Non governmental organisation
NRU Nutrition Rehabilitation Unit
ODI Overseas Development Institute
OTP Outpatient Therapeutic Programme
RC Relief Committee
SADC Southern Africa Development Community
SC UK Save the Children UK
UNICEF United Nation Children's Fund
UNHCR United Nations High Commissioner for Refugees
USAID US Agency for International Development
WFP World Food Programme

Scope of the supplement

People fleeing during crisis in Indonesia

This supplement is intended to provide guidance on the design of food targeting systems in emergencies. Targeting is defined as directing a particular type or quantity of food, to a defined population group. A broad definition of emergency contexts has been used, to include rapid and slow onset emergencies and responses aimed at emergency preparedness, in acute and protracted settings.

Developing targeting systems, which can be operated effectively at reasonable financial and administrative cost, has been a focus of work in the humanitarian and development sectors in recent years (Barrett, 2002):

"The key alleged problems surrounding food aid - displaced international trade, depressed producer prices in recipient countries, labour supply disincentives, delivery delays, misuse by intermediaries, diversion to resale or feeding livestock or alcohol brewing, dependency, inattention to beneficiaries' micronutrient needs, etc. - all revolve ultimately around questions of targeting. If the donor community could improve the targeting of food aid, it could improve the effectiveness of food aid in accomplishing its primary humanitarian and development aim - the maintenance of valuable human capital - and reduce many of the errors that sometimes make food aid controversial, ineffective, or both." (Barrett, 2002)

The move towards greater accountability for food aid comes in the context of a recent decline in global food aid availability (in 1991 there were 12.21 million MT global food aid in the form of cereals, compared to only 7.35 million MT in 20011) and an increase in the number of protracted emergency contexts (in 2002, 17% of WFP food shipments went to protracted emergency contexts) with a concomitant pressure to phase-out food aid distributions.

While there have been some significant developments in food security assessment techniques and in targeting practice, notably a growing body of experience of community based and managed targeting, many practical difficulties remain. Food aid donors sometimes impose their own targeting objectives, e.g. that food should be given only to children under a particular malnutrition threshold or to female headed households. Existing, sometimes weak, administrative structures may be the only practical method of targeting food to large populations. Many NGOs may be involved, each with its own views on targeting, e.g. Save the Children is more likely to target children and HelpAge International, older people, often with little overall co-ordination. Food aid may be more or less available in different locations and the availability of food aid or particular commodities may change over time for extraneous political and bureaucratic reasons (Clay, 2000).

Selling maize husks in Malawi

These constraints can rarely be influenced by any single humanitarian agency. Nevertheless, there is sometimes scope for significant targeting improvements both in large-scale operations and at a local level within larger food operations. The aim here is to lay the basis for a logical approach to targeting, given a particular set of constraints.

Geographical targeting, i.e. distributing food to the population of one geographical area and excluding another, is not discussed in any detail in this supplement. All food aid is geographically targeted. Targeting may take place between countries, between regions in a country or between sub-region, food economy or livelihood zones, districts or villages. These targeting decisions are usually informed by broad sets of indicators, which determine the extent to which different areas are affected by a given shock. This supplement concentrates on targeting decisions that are made after the geographical targeting decisions have been made. In many situations, geographical targeting provides the best strategy for achieving the targeting objective and further targeting may be unnecessary.

Targeting within geographic areas is appropriate when:

  1. There are identifiable differences between the intended target and non-target population
  2. The targeted population is a minority of the population
  3. It is operationally feasible to implement a targeted distribution
  4. The community co-operates with the targeting strategy

(Sharp, 2001)

There are many situations where food aid is not the most appropriate intervention and where other types of intervention are more suitable to meet people's needs, e.g. cash distribution, livelihood interventions and food price support. These are not covered in detail, as they raise a wider set of technical and political issues and require more extensive discussion than is possible here. Methods for determining food needs, the detail of supplementary feeding and other food outlets, logistics, administration and the practical implementation of targeting strategies are highlighted only where these are directly relevant.

Lastly, there are situations where food resources are grossly inadequate relative to needs. A special section is devoted to these contexts.

Why target food?

There are four reasons for targeting:

  1. To ensure food aid is received on the basis of need.2
    Assessments often show that people have different degrees of need in an emergency. Targeting is often used as a mechanism to reach those in greatest need.
  2. To avoid harm.
    Targeting can reduce the quantity of food supplied and reduces the risk of depressing producer prices and production, disrupting trade, or displacing traditional social reciprocity networks.
  3. The efficient and effective use of resources.
    Food aid is sometimes regarded as a 'free' resource. In fact, the large transport and other associated costs do often compete with other non-emergency investments. Where the targeted population is a minority, accurate targeting can reduce costs. Targeting can be used to maximise the impact of a given quantity of food.
  4. Insufficient relief food.
    Early warning failure, political, logistical and security constraints may lead to late and inadequate deliveries of food, and decisions must be taken about who is to receive this.

Basic concepts and definitions

Eligibility criteria The characteristics of individuals or groups e.g. households, who are eligible to receive food.

Coverage The proportion of the eligible population that actually receives the intended ration. The ease and accuracy with which coverage can be measured depends on how easy it is to identify those who are eligible.

Exclusion error The proportion of individuals eligible to receive food but not accessing it. Assuming the overall size of the targeted population is known, the exclusion error can be derived from the estimate of coverage.

Inclusion error The proportion of individuals accessing the food who are not eligible to receive it (see Figure 1).

Terminology 'Household' has been used to mean a group of individuals, usually related, who form an economic unit. At the extremes, a household might be a single individual or one hundred or more. 'Community' is taken to mean a group of households that are, to a greater or lesser extent, economically interdependent, e.g. a village. It would, for example, typically exclude a refugee camp, at least in the early stages of its existence.


Overview: steps in planning a targeting system

The aim of a targeting system is to use the available set of resources (food, skills, and cash) to meet the needs of a defined population as effectively and efficiently as possible, i.e. to maximise coverage and to minimise inclusion errors.

Targeting systems are developed in a vast array of situations. Areas may vary in terms of population density, level of infrastructure and communications, social and government organisation and degree of security. The broad pattern of targeting may be dictated by Government and/or food aid donors, sometimes in terms unrelated to the realities of the location. There may be few or many external agencies, each with its own objectives, and large or small quantities of food available for distribution. The intervention may be early, and based on an adequate assessment, or ad hoc and after starvation has occurred.

A targeting system is, therefore, unique to each situation and it is not possible to lay down hard and fast rules. Targeting is a pragmatic exercise requiring judgement, compromise and, in some situations, active evaluation and the modification of strategy as the situation develops. "Perfect targeting is an impossible ideal. The best that programme designers can hope to achieve is to reduce targeting errors to acceptable levels" (Devereaux, 2000).

The approach suggested here is to work systematically through the steps involved in targeting, at each step looking for the potential errors which might occur, and modifying the approach accordingly where this is practical. Errors will inevitably occur at each step in the process; the accuracy of the overall system will be the sum of all the errors.

It is important to understand that targeting is a process rather than a defined activity and it, therefore, relates to all aspects of the project cycle. There are five interrelated steps in designing a targeting system:

Assessing and defining needs (Section 1)

There is no absolute or universally accepted definition of food needs. In the same location, needs might be defined differently by different agencies. For example, a Government might wish to ensure that people had enough food in the short run and could maintain their ability to provide for themselves in the long run, e.g. would not have to sell livestock and other assets to survive. A donor might wish to provide only for immediate household needs and to provide food specifically to women. An NGO might wish to provide food only to malnourished children.

Affected people also have a view of need that may conflict with that of the donor or agency distributing food. In many emergencies, most people may be chiefly preoccupied with preserving their assets and securing their means of subsistence in the long term, rather than with the risk of starvation. Recipients may also have views of 'entitlement' that fundamentally differ to those of food donors, e.g. they may consider that people displaced into an area or of a different ethnic group are not entitled to food aid.

Setting objectives (Section 2)

Targeting objectives arise from the assessed or assumed food needs of an affected population. The targeting objective determines the design of a targeting system, i.e. it defines the groups who should receive food, the quantity and quality of food they receive, when they should receive it, and why. Objectives need to take into account errors that occur in assessment and errors that are anticipated in targeting.

Determining eligibility to receive food (Section 3)

Determining eligibility has two elements, the first is setting the eligibility criteria, and the second, applying the criteria in practice. The eligibility criteria arise from the targeting objective. Eligibility criteria may apply to individuals, e.g. children less than -2 z scores weight for height, or to households, e.g. those with less than 0.5 hectare land. Eligibility criteria must be defined in such a way that they include the intended beneficiaries and can be used in a practical way to identify individuals or households to receive food.

Eligibility criteria may be set by external agencies (administrative targeting), defined by the community (community managed targeting), or an accommodation may be reached between the two. Where donor and community based views of need conflict, communities often have considerable scope to subvert externally imposed targeting objectives.

The application of the criteria can be done by the agency delivering the food or by the community. In practice, the identification of targeted households is usually done by the community or its administration, e.g. the chief. The targeting of individuals can more easily be done by the external agency

Choosing a method for distributing food (Section 4)

Food can be distributed in many ways. Distribution may be through existing facilities, e.g. health centres, schools, or through agency designed and controlled centres, e.g. supplementary feeding centres. Food may be provided directly to households through home-based care or given directly to the community in bulk for redistribution.

A distribution method must be chosen that maximises coverage of targeted people and minimises inclusion errors. In general:

  • Coverage will be increased by minimising the cost to the beneficiaries of using the food distribution system, i.e. travelling distances, distribution frequency and waiting times. With dispersed rural populations, these criteria are most easily met using decentralised, community based distribution methods.
  • Inclusion errors result when the non-targeted population are included. This can result from diversion at various levels including bias, corruption, pressure, theft by various parties, and by redistribution and sharing within households. Inclusion errors can be reduced by tightly controlling the distribution process, as in centre based feeding, or, where feasible, by handing over responsibility for distributions to the community. Involving women in the distribution of food can be an important part of reducing inclusion errors.

Developing a monitoring system (Section 5)

Monitoring should determine whether the objectives are being met, thereby allowing the targeting system to be redesigned and improved if necessary. Monitoring can be conducted to measure errors occurring in assessing and defining needs, setting objectives, determining eligibility and in choosing a distribution method. There are multiple methods which can be used for monitoring and evaluation. Data from several sources should be triangulated in order to gather a complete picture of the effectiveness of the system.

In designing a targeting system, a practical balance must be struck between:

  • Imposing eligibility criteria that may be ignored and reversed by the recipient communities, and
  • Imposing controls on food distribution, which effectively exclude needy people either a) because eligibility criteria are too restrictive, e.g. targeting malnourished children, where adolescents and adults are also in distress, or b) by providing food in a way which makes it difficult or impossible for people to acquire the food, e.g. through very centralised systems.

In practice, the precision with which food can be targeted will tend to vary with:

  • The quality of the information available.
  • The ability to reach a practical working arrangement with the community that meets both externally perceived needs and those recognised by the recipient population. It is generally easier to do this with more physically accessible and smaller populations, where there can be good contact between the external agency and the recipient community. However, much depends on local circumstances, e.g. whether people are settled or mobile.
  • The quantity of food available for distribution. The larger the available food aid relative to the perceived needs of recipients, the easier it is to accommodate both donor and community objectives. In settings where information is inadequate and/or the food available is less than the assessed need, targeting may break down entirely. Under these conditions, systems may have to be adopted which impose criteria to ensure that the minimum survival needs of the maximum number of needy people are met (see Section 6).

In some settings, e.g. where large amounts of food are available for small populations, or a majority of the population is in need, the costs of targeting may exceed the value of doing so. The cost considerations of different targeting systems are described in Section 6.


Table 1: Overview of the targeting supplement
Steps to design a targeting strategy Aspects covered in the supplement Influencing factors
1. ASSESS & DEFINE NEEDS Needs can be defined according to
. economic
. vulnerability, or
. outcome assessments
. Donor agency priorities
. Implementing agency mandate
. Community perspectives
2. SET OBJECTIVES Who should get how much food aid and when and why? e.g.
. Life saving
. Asset protection
. Quantity and quality of food aid available
. Timing of the response
. Resources available in addition to food aid

. Malnourished
. Pregnant and lactating
. Ill
. Elderly
. School children

. Socio-economic, including gender, HIV/AIDS affected, asset holdings or income/ employment
. According to nutritional status of household members

Self targeting:
. market
. commodity choice
. food for work

Ability to actually identify the eligible individuals or households using the criteria.

Both the suitability of the criteria and the task of identification need to be considered.

4. DISTRIBUTE FOOD . Maximise coverage
. Minimise inclusion errors
. Respect human rights principles
. Context (conflict, displaced, large area, etc)
. Systems already in place, history of targeting
5. MONITOR SYSTEM Types of monitoring include:
. Food basket monitoring
. Post distribution monitoring
. Food usage and market surveys
. Impact evaluation
A system is never perfect

1Total food aid available each year is highly dependent on international commodity prices. IFRC and ICRC, 1994.

2Non-discrimination and impartiality are cornerstones of humanitarian law. This means that food aid must always be distributed according to need (rather than meaning everyone needs to get an equal share). The Code of Conduct for the International Red Cross and Crescent Movement and NGOs in disaster relief (1994) states that "Aid is given regardless of the race, creed or nationality of the recipients and without adverse distinction of any kind" and "Aid priorities are calculated on the basis of need alone".

Imported from FEX website


About This Article

Article type: 
Special section

Download & Citation

Recommended Citation
Citation Tools