Development reasons

:  so that dependency and economic disincentive effects

are minimized.

Conceptually, the basic definition of and rationale for targeting is clear.  In practice, however,

targeting is extraordinarily difficult and perfect targeting systems are virtually impossible to

implement, even in countries with well developed information and delivery systems.  Errors of

exclusion or inclusion must be accepted, although the goal of any targeting system must be to

reduce both.   Beyond the basic definition of targeting, there is no universally agreed upon

terminology to define more detailed concepts of targeting

From an operational point of view, the questions, which those involved in allocating food aid on an

objective basis want to know, are:

1) WHERE is vulnerability to drought (flood) the greatest,

2) WHO are the groups most vulnerable to this risk,

3)  WHAT  type of (food aid) assistance is needed,

4) WHEN is it needed, and

5) HOW can it be provided in order to ensure that food insecurity is reduced, development

mechanisms are reinforced, and that the intervention is cost effective.

It is clear that the effectiveness of targeting strategies will improve the greater the extent to which

information is available to answer these questions.   The conceptual approach should reflect the

need to address these basic questions.

Two concepts that provide one organizing structure for addressing these questions are the LEVEL

of targeting and the MECHANISM used to target. The LEVEL of targeting primarily relates to the

WHERE and the WHO, whereas the MECHANISMS relate to the HOW.   Answers to the WHAT

and the WHEN also influence targeting design decisions specifically related to the level and

mechanisms used to target.

Targeting can occur at many different LEVELS.  Donors may target food aid resources to one

country over another; governments, donors or NGOs may target food aid to one administrative

region over another; local authorities may target food aid to one household over another; and a

household may target food to one individual member over another.  In general, however, most

targeting decisions focus around the middle two levels, commonly known as area targeting or

household targeting.

There has been a long standing debate about whether resources are best spent on improving area

level targeting (also known as geographic targeting), or on household targeting.  Theoretically, this

depends on the homogeneity of the unit.  When inequality within areas or communities is less than

inequality between areas, greater gains can be made in impact and efficiency by discriminating

between areas than between households (Sharp, 1997).  A recent study in Ethiopia

23 

strongly

argued that resources would be best spent on area targeting.  That study recommended that

 more emphasis should be placed on identifying the most food insecure 

weredas

 (area targeting)

as the first step in the food aid targeting process.  Efficient area targeting has a greater likelihood

of reaching vulnerable households, and possibly at lower cost, than does household level

targeting.   Because the efficiency of area targeting depends on the homogeneity of a region, the

smaller the administrative region of analysis, the more likely the inhabitants will be homogeneous.

Many different MECHANISMS can be used to target at various levels, including market targeting,

demographic targeting, self targeting and administrative targeting.  In reality, a combination of

mechanisms, employed at the same or different levels, often ensures the most successful

targeting.  A basic consideration is degree of control over the targeting process; namely the degree

23 

Clay, D., Daniel Molla and Debebe Habtewold, 1998, Food Aid Targeting in Ethiopia:  A Study of

Household Food Insecurity and Food Aid Distributions, Working Paper 12, Grain Market Research Project,

Ministry of Economic Development and Cooperation, Addis Ababa, Ethiopia.

A 71