Evaluation of the Wajir Relief Programme 1996-98

Published Evaluation

OXFAM (UK and Ireland) were engaged in relief work in Wajir District of Kenya from September 1996 until October 1998. The interventions were a response to the droughts of 1996/7 and the El Niño floods of 1998. OXFAM published the findings of an external evaluation¹ of their response in April 1999. This evaluation drew on the experiences of their staff, other agency and government staff and on the views of beneficiaries.

Wajir district is populated by a largely pastoralist population. Successive droughts had led to depletion of livestock herds and increasing destitution. Available data on child malnutrition demonstrated a severe crisis. OXFAM responded by organising one of the largest relief operations ever mounted in Wajir District. Over the two year period OXFAM spent £4,612,029 on relief components. It delivered over 21,000 tons of maize 16,000 tonnes of beans and 500 tonnes of oil.

Although the programme was continuous, it can be broken down into two distinct phases:

1. The response to the drought starting with a three month drought mitigation project in the south and parts of the west extending to the whole district except a 15 km radius around Wajir town (where the Kenyan government were providing relief) and winding down by October 1997;
2. The response to the El Niño floods with a relief programme from December 1997 to August 1998 which became a recovery programme ending in October 1998.

Overall the evaluation demonstrated the timeliness and high quality of the relief response delivered by OXFAM to the rural population of Wajir during the drought and floods of 1996-8. However, there were problems:

1. there was a tendency to plan and submit proposals for only 3 months. This appears to have been due to a reluctance by DfID to commit to longer term food purchase as their preferred option was to encourage WFP to provide relief food;
2. the planning phase consistently underestimated the number of beneficiaries eligible for relief food. As a result the ration had to be cut and the food shared more thinly.

Main findings of the evaluation

OXFAM used a system of relief food distribution whereby the district was divided into five livelihood zones, each with a different recovery package. This approach was appropriate and worked well.

OXFAM registered women as beneficiaries of relief resources. Although more management intensive than other distribution approaches, it was a price worth paying for a more efficient and equitable means of distribution.

The relief food became a critical component of the diet and the dramatic fall in malnutrition rates during 1997 over a period of about five months was probably due to the combined results of the relief operation and of good rains in April/May 1997.

The food distribution also stabilised food prices helping pastoralist purchasing power. This benefit was especially critical for Wajir town which was less well served by the relief operations due to smaller distributions of maize by the GoK. This presented a dilemma for OXFAM as many of the poorest and most vulnerable lived in the town but it would not have been appropriate to shoulder the governments responsibility and replace their relief operations with an NGO one.

The coverage of 90% of households reported in July 1997 was exceptional and the operation was consistently rated by beneficiaries as one of the most successful the district had experienced. The impact of the floods was sudden and extreme. During the first few months of 1998 when the road network was severely disrupted, the food distribution was life-saving. OXFAM supplemented the meagre government relief ration in Wajir town with Unimix and high energy biscuits.

As Wajir district began to recover from drought and flood OXFAM altered their approach and instigated a series of recovery interventions to assist vulnerable households as food relief was withdrawn. A financial analysis of the four types of intervention demonstrated the costs and benefits associated with each. It cost Ksh 525 to deliver one ration of food relief. In order to provide similar benefits, i.e. the equivalent of one monthly ration, restocking with goats cost Ksh 712, restocking with oxen Ksh 480 and cash for work Ksh 450.

The cash for work scheme made the most impact, allowing households to clear their debts, buy essential items and invest in small-scale income generating activities.

Restocking interventions also provided benefits to recipients but sheep and goats did not achieve the levels of production anticipated. The result was that this became an expensive means of transferring assets to poor households. Restocking with oxen and provision of seeds would have made a significant impact if the short rains of 1998 had not failed.

The on-going development programme undoubtedly suffered from the legacy of the largescale relief programme. It dramatically changed the perception of OXFAM in the district, which now tends to be seen as a 'free hand out resource rich' relief agency. OXFAM's ability to turn the image around and re-establish itself as a development agency will depend partly on the time lag before another emergency operation.

OXFAM exploited its comparative advantage as an international NGO with strong field based operations by identifying and responding early to localised food stress and mobilising fast as the emergency developed and changed. However, the success of its response was highly dependent on funding provided by one bilateral donor -DfID. The lengthy procedures that WFP must follow to launch an appeal and deliver food to Kenya meant that the first WFP deliveries to Wajir arrived 9 months after OXFAM had started the relief operation and four months after the GoKs emergency declaration.

¹Buchanan-Smith. M and Barton.D (1999); Evaluation of the Wajir Relief Programme (1996-98). Oxfam, Oxford OX2 7DZ, England.