only received a small percentage of their requirement   [FAO/WFP 1996, ref 102].

Several government officers at different levels, and some village informants, mentioned the

assumption that free distributions are self targeting: that is, whoever turns up at the distribution site

when a distribution has been announced is assumed to be in need,  because the better off would

be ashamed to come and claim relief.   From conversations in the villages it does seem to be true

that a small number of higher status people had voluntarily excluded themselves from claiming

relief, but this is clearly not a sufficient targeting mechanism to ensure a worthwhile impact on the

really needy.

Another observation confirmed by field observers in many areas was that the quantities (and

frequency) of food deliveries received at community level were often very much smaller than the

allocations on paper in Nairobi.  This is partly accounted for by the practice of paying transport and

handling costs out of the maize allocations (as there is no separate provision of funds for this).

However, leakage at various levels of the system seems to be more significant than can be

accounted for by this alone.

An important point is that blanket distribution at the final point of delivery to beneficiaries (i.e. lack

of Stage 3 targeting), while highly visible to field observers, is only a part of the problem.  Tighter

accountability to prevent leakage, and more ruthless selection of the neediest Districts, Divisions

and Locations (i.e. area targeting at Stages 1 and 2) are equally important if the GoK wants to

improve the impact of its relief distributions on its food insecure citizens.

3.2.c.  Focus area 1:  Mwingi District

Mwingi, in the Eastern Province, is a marginal agricultural area which was highlighted by FEWS in

late 1996 and early 1997 as one of the areas severely affected by three consecutive harvest

failures.  Mwingi was not included in the WFP drought EMOP, but was allocated significant

quantities of government relief maize (see Table 12) and was also supported by the GTZ

Integrated Food Security Project (IFSP E) which has been established in the District since 1994.

The study team focused on two main issues here:   the targeting of community Food for Work as a

potential alternative to free distributions, and local government perspectives on the targeting of

free relief within the District.

Intra District  targeting of government free relief

District officials explained that government relief food is distributed through Social Dimensions of

Development Committees at District, Division and Location levels.  Membership of these

committees includes the heads of government departments, NGO representatives, churches and

politicians.  The DSDDC listed six information sources available to them in allocating food aid to

Divisions:

1.  Reports from Location Chiefs on the numbers of people needing food;

2.  Information on rainfall, crop conditions, livestock etc. through the District Agriculture Office

from their officers at Location level (the DAO produces a monthly food situation report,

including a District food balance sheet );

3.  Nutritional status information from Location dispensaries, through the District Health Office;

4.  Reports from NGOs on the areas where they work;

5.  Politicians requesting food for suffering people in their area; and

6.  The Drought Monitoring System, which has been established in Mwingi with GTZ support

since 1996 (closely modeled on the DPIRP / ALRMP system outlined in section 3.2.a.). The

Drought Monitoring System, managed by the District Statistical Officer, also produces  monthly

reports.

As at national level, the problem does not appear to be lack of information (at least in this District).

During the drought crisis, the DSDDC met monthly and, putting these sources of information

A 41