SAVE THE CHILDREN DENMARK
SOMALILAND PROGRAMME

BY AHMED OSMAN IBRAHIM
DECEMBER 2004

ABBREVIATIONS

CLHE Candlelight for Health, Education and Environment
- CBFTDM Community-Based Food Targeting, Distribution and Management
- CE Community Elders
- CBO Community-Based Organization
- CS Coping Strategy
- DANIDA Danish International Development Agency
- DDC District Drought Committee
- DKK Danish Kroner
- FM Food Monitor
- FS Field Supervisor
- FSAU Food Security Analysis Unit
- HH(s) Household(s)
- HOVOYOCO Horn of Africa Voluntary Youth Committee
- HQ Headquarter
- INGO International Non-Governmental Organization
- LOA Letter of Agreement
- LFW Logical Framework
- NERAD National Environmental Research and Disaster Preparedness Authority
- PM Programme Manager
- PRA Participatory Rural Appraisal
- RG Regional Governor
- RDC Regional Drought Committee
- RPC Relief Project Coordinator
- SCD Save the Children Denmark
- SCA Save the Children Alliance SC-UK Save the Children UK SC-USA Save the Children US
- FFW Food for Work
- MT Metric Ton
- NGO Non-Organizational Organization
- SSI Semi Structural Interview
- TOR Terms of Reference
- TOT Terms of Trade
- WFP World Food Programme
- VC Village Committee

0. EXECUTIVE SUMMARY

1. The drought situation in eastern regions of Somaliland was gradually getting worse since 2001 spreading to other areas of the country. Following government humanitarian appeal in October 2003, interagency assessment was first conducted in December 2003 and later nutritional surveys. The assessment confirmed the death of 40-50% shoats, 15-20% camels and 95% of cattle, the only livelihoods of the drought affected pastoral communities. As result, the affected pastoral communities could not meet and sustain their needs and livelihoods. The assessment therefore recommended immediate interventions in food and water among others to avert heavy loss of lives and livelihoods.

2. As part of the humanitarian responses, SCD Somaliland programme developed a proposal based on the assessment findings in December 2003 and submitted to its Headquarter for consideration. Having analysed and approved the proposal of child focused emergency food distribution, the SCD HQ approached DANIDA for funding. Fortunately, DANIDA in mid January 2004 approved the project with a total budget of 3,000,000 DKK. SCD and DANIDA have reacted positively and responsibly to the emergency project with immediate approval of funds. This has considerably contributed to relevance, timeliness and effectiveness of the project.

3. SCD Somaliland Programme Office implemented a Child Focused Emergency Project in Togdheer region of Somaliland in 2004. The goal of the project was to contribute to the improvement of the food security situation of the affected people in the Togdheer region of Somaliland. The project purpose was to create an improved situation for survival, development and protection of 30,000 people of which 6000 are children within in the targeted villages of Togdheer region of Somaliland. To achieve the project purposes, three outputs have been planned: access to basic food items provided for beneficiaries for 3 months from March to May 2004, access to educational materials support provided to 1500 children selected from 15 schools of Togdheer region and strengthen capacity of the disaster committee including community committees through training.

4. At the end of the project, two consultants have been hired to do the final evaluation. The main objectives of this evaluation were: to examine the extent to project objectives were met, the effectiveness of the project in reaching the intended or right target groups especially children, the impacts or the changes the project brought up in improving the situation of the affected beneficiaries; and to assess the household food economy/distribution system and provide concrete recommendations.

5. The Evaluation involved all stakeholders at different levels including those from Hargeisa, Burao of Togdheer region, and community levels. Views of some possible key informants were taken. Relevant documents have been reviewed before actual data collection. Semi-structured interview, focus group discussions, general public gathering and discussions were the main methods of data collection. Village Committees (VCs), women groups, teachers, children and community as well as beneficiary households in each of the visited 6 communities have been interviewed. Discussions have been undertaken with agencies and key informants such as the SCD staff in Hargeisa, project field staff in Togdheer region, WFP, HOVOYOCO, CLHE, WFP, OXFAM GB, SC-UK, NERAD, RDC and RG etc.

6. The evaluation focused on all aspects of the project from the designing, planning, implementation, monitoring and achievement as well as impact of the project. It revealed valuable information on those areas, which will contribute to the improvement of similar future projects. At end of the evaluation exercise, two separate reports one on evaluation and the other on household food economy have been prepared.

7. The project introduced participatory approach of food distribution and management. The planned implementation period was between March and May for three months but was extended to October as SCD and WFP agreed to distribute food alternatively in the target areas on monthly basis. Recruitment of the project staff and other project preparatory works have been carried out in February 2004. Two days of training on targeting from 3-4 march 2004 have been conducted in Burao, the regional capital of Togdheer in which 7 staff and 2 members from regional office attended. Unfortunately, no members from the target communities have benefited from the training.

8. As soon as the training has been finalized, initial assessment, community mobilization, registrations and targeting that continued for 15 days had started simultaneously in all target areas. Initial assessment and community mobilizations have been carried out in 41 villages. It was not possible to cover other villages due to limited number of projects staff (7 persons). The targeting and registrations based on nutritional screening have generally gone well in the first distribution as the most vulnerable groups left in the villages such as children and women have been identified, registered and targeted. The project provided identification cards to the registered vulnerable households. All registered households (HHs) received the full-allocated food. Project targeting was relatively better than that of WFP which community called `sit on the sack' meaning every person in the village even the travellers could get food if they meet the distribution by chance. Nevertheless, in-migration and out-migration of pastoral population, deterioration of the drought situations and underestimation of the population from the interagency assessment have reduced effectiveness of registration and retargeting in the second and third distributions. Inclusion and exclusion errors have been reported in the target communities.

9. In general terms, the distribution was considered fair and responsive, transparent, and appropriate to local conditions. All registered HHs received the food. Proximity of the distribution sites to the villages has made the beneficiaries happy. The announcement of the distribution time well before the actual distribution is consistent to Sphere Standards. The project ensured that the food reached the family through women, as most of the recipients who physically collected food were women. The most vulnerable groups have been given the first priority. In fact, SCD food distribution project was the best compared with other projects in the country in general and the affected areas in particular because of provision of locally appropriate food with relatively adequate quantity as well as timeliness of food distribution. The communities have the maximum respect for SCD and would rank it as first closest organization if PRA institution analysis were conducted in the region.

However, as the distribution is closely linked to targeting, the same factors affected the distribution negatively. Because of the increasing vulnerability of the relatively better off groups and unregistered needy households, the community and beneficiaries agreed to share with other needy households as sharing is a norm. In some cases, all households in some of the target villages received part of the food especially in the second and third distributions. The quantity of food provided consisting of 50 kg of rice, 10 kg of sugar and 3.65 kg of vegetable oils per 6 member households per month was close to the Sphere minimum standards though deficient in proteins and other essential micronutrients. The food could be adequate for 3 weeks instead of one month.

10. Although the project covered the needs of some of the most vulnerable groups within the target communities satisfactorily, the needs were high in the target areas as compared to project inputs. This is because the situation seemed to have been worsening from time to time and many people considered as better off initially slipped down to the most vulnerable category to the extent that there was no difference between the two groups. Unregistered needy households who moved at the beginning came back to their original settlements. This has consequently decreased the effectiveness of project coverage.

11. The project was highly relevant to the needs of the community as it addressed the priority need of communities (food). The communities have been also consulted on the food choices and food basket. The project provided the most appropriate food, the common food basket in the region. The distributed food has been exclusively consumed at the household. Apart from the sharing with needy relatives, there was no exchange of food or sale of food reported in any of the communities. When asked the reason, they mentioned that what they needed was only these and there was nothing better to exchange with the required food. The community also mentioned that the project inputs such as distribution of food started at a time when they had exhausted their livelihood base and were on verge of starvation.

12. The SCD project efficiently used human, material and financial resources at its disposal. Human resources and transport have been hired during targeting and distribution and terminated as soon as the workload declined. All the financial resources have fully been utilized to address the needs of the most affected communities. The project distributed 745MT of rice, 124.2 MT of sugar and 45.75 MT of vegetable oil to 5000 HHs. The plan was to distribute 750 MT of rice, 150 MT of sugar and 54 MT of vegetable oil. However, the planned quantities were reduced due to inflation of food commodity prices. SCD carried out regular and timely budget monitoring whereby surplus budget amount was used to make additional food purchase and distribution to 22 communities, after the approval of no cost extension by DANIDA. The planned costs of the project are almost similar to the actual cost at the end of the project period.

13. The project coordination and collaboration with other agencies and local authorities was satisfactory. Weekly and monthly coordination meetings have been arranged at national and regional levels at the onset of interventions. Geographical target locations and sectors of intervention have been divided. Roles and responsibilities of different organizations were clearly defined. Information on progress and constraints were shared in the different level coordination meetings thought coordination meetings later became nominal. However, it was noticed that there exists difference among aid agencies in terms of targeting of beneficiaries, type of food, quantity and quality of food to be distributed in the target.

14. The project has carried out most of the planned activities. The first distribution took place in March, while second and third distributions at one go have been carried out at the end of May and first week of June 2004.The most accomplished activities that could fall under the first and third outputs have led to the achievement of the most important project outputs `access to basic food provided to the target beneficiaries'' and `strengthened capacity of disaster committee including community committees through training'. However, the second project output `access to educational materials support provided to the 1500 children selected from 15 schools of the region'' implemented through SC-UK is not fully achieved as some of the visited communities have not verified distribution of educational materials. SCUK asserted that most of the educational materials were provided to schools to be used by teachers and students at school level.

15. Having achieved the purpose of the project, the food distribution averted the eminence of large scale of famine by decreasing hunger, starvation and related diseases and ultimately heavy loss of lives. It enabled the target group to survive and reduced worry and frustration of food crisis. It also saved livelihood base as selling of remaining few breeding animals for food have been spared. The communities also mentioned that purchase of food from the market on credit or by any other means declined as appropriate food became available. It has also reduced begging of food by vulnerable groups as they received food and decreased the burden of the community on collecting food contribution to the needy and vulnerable households.

16. The project had also a significant impact on the child education. It enabled continuation of child education in the drought-affected community. School children and their families received their share of the food.

Furthermore, the food distribution also reduced further migration either to towns or to other villages in search of food, pasture and water. The project additionally brought about social benefits to the recipient households and reduced divorce of the existing families as result of lack of food and income to satisfy the their needs and increased marriage. In addition, it contributed to the improvement of nutrition and reduced deteriorating situation of malnutrition.

17. The suggestions for future improvement include proper beneficiary targeting by raising the awareness of the community on emergency food aid, facilitating election of gender and children sensitive food distribution community committees, capacity building of the committees on targeting, gender issues, monitoring and management of food distribution, setting up of targeting criteria with the communities using PRA wealth ranking, wide dissemination and approval of the criteria at public gathering, updating and approval of the list of targeted beneficiaries and re-planning of the project inputs based on joint initial assessment as well as avoiding the fixed quota.

18. The project is also recommended to increase the coverage of vulnerable groups as most of them lost their livelihood base. The number of targeted beneficiaries is to be increased to 7500-10,000 households depending on resource availability. The project should have separate target areas to avoid community confusion on the approach of food targeting and distribution, type of food, quantity and quality of food as well as timing of distribution of different agencies.

19. Monitoring system capturing the key issues in the project to be put in place before implementation is essential. Both monitoring for efficiency and effectiveness should be carried out in the project lifespan. The staff and the committee should be involved in the monitoring.

20. The project is recommended to increase the level of community participation. Decision-making process of targeting, registration and distribution, monitoring and management of food distribution should be in the hands of the community. The staff should take the facilitating role.

21. Of the recommendation include provision of 75kg of rice, 20kg of dried beans, 7.2 kg of vegetable oil/household/month. This will meet the sphere minimum standards in terms of energy, protein and fats though still deficient in some other essential micronutrients, which the target communities could gain from other sources. The project should maintain meeting of Sphere standards and encourage others to do so, and continue food distribution up to June 2005 as the dry period (Jilal) is to start where food security situation in drought affected pastoral community is not expected to improve in the near future

22. The project is recommended to formulate an exit strategy and design in the project proposal. SCD should link up with those organizations, which are planning drought recovery programmes such as Oxfam GB, PENHA, HOVOYOCO, CLHE, and sort out the issue before the next rainy season in May 2005.

1. INTRODUCTION

Save the children -Denmark Somaliland Programme implemented a child focused emergency food aid distribution project in Togdheer region of Somaliland in 2004. The project introduced participatory approach of food distribution and management. The planned implementation period was between March and May for three months but was extended to October as SCD and WFP agreed to distribute food alternatively in the target areas on monthly basis. During this project period, 745 MT of rice, 124.2 MT of Sugar and 45.33 MT of vegetable oil have been distributed to 5000 households' equivalent to 30,000 beneficiaries of whom 6000 were children. At the end of the project, two consultants have been hired to do the final evaluation. The three objectives of this evaluation as described in the Terms of Reference, were:

1) To examine the extent to project objective were met, the effectiveness of the project in reaching the intended or right target groups especially children, the impacts or the changes the project brought up in improving the situation of the affected beneficiaries.
2) To assess the household food economy/distribution system. This included the understanding of the sources of food and types of food the affected people secure in addition to relief aid, the decision making responsibility in the household food economy, the adequacy of food aid assistance in fulfilling the food gap of the households based on sphere standards, criteria used in the household for distributing food among members of the family, feeding practices at household level etc.
3) To forward concrete recommendations based on the evaluation findings and suggest the food basket needed for the household based on sphere standards, system of monitoring of the food aid assistance in order to ensure that it reached the intended beneficiaries especially children and exist strategy in emergency operations.

Based on the Terms of Reference see annex 2, the two consultants carried out the evaluation for 15 effective days from 26 November to 12 December 2004. The consultants were accompanied by one of the project field supervisors. Evaluation has incorporated information from relevant documents and from all concerned organizations such as HOVOYOCO, CLHE, WFP, OXFAM GB, SC-UK, NERAD, RDC and RG. The consultants also visited 6 of the 41 target villages and had discussions with village committees, communities at public gatherings, interviewed beneficiaries of women, men and children in each of the visited communities. This consultancy report is the result of this participatory evaluation exercise. It is organized in five sections including this introduction as section one. Section two gives the background to the project, the socio-economic situation of the area, planning and resource mobilization and the content of the project. Section three deals with the evaluation itself, highlighting the purpose, methodology and the strengths/constraints of the evaluation. Section four discusses the findings of the evaluation. Under this section, seven sub-topics like project design and planning, pre-implementation activities, implementation, achievements and impact, resource mobilization and utilization, coverage, coherence (collaboration and coordination) and monitoring are addressed. The last section, section five provides the conclusions and recommendations. In addition four annexes are attached for further information.

2. BACKGROUND

2.1 Socio-economic situation of the area

Togdheer is located in the middle of the country bordering Ethiopia in the south and Sool in the East, Sanaag in northeast and Galbeed region in the West. The region is characterized by chronic food insecurity with considerable high malnutrition rate. Periodic drought and food insecurity have been recurring to the region since 1960s. Famine situations or food shortages have been recorded in the periods of 1964, and 1974. Last year, 2003, the whole region was declared food deficit due to lack of rains for three consecutive years. As a result in July 2003, the government of Somaliland made appeal to the international community for assistance.

Togdheer with an estimated population of 350,000 is heavily dependent on livestock and more than 70% of the population is pastoralists. The region is divided into five districts namely, Burao the regional capital, Oodweyne, Qoryaale, Balli-dhiig, and Duruqsi. Animal husbandry is the main source of the economy in region. The majority of the people depend on livestock mainly camels, goats and sheep for their living. Income from sale of charcoal, labour and firewood also contributes to household food security. Heavy environmental degradation is one of the basic causes of the recurrent drought in the region. The situation was exacerbated by the livestock ban imposed on the country by the Gulf states particularly the Saudi-Arabia, the biggest market destination of Somali livestock. The continuous asset depletion and unfavourable TOT have weakened the purchasing power of the people. Thus, though food was available in the market in towns, people could not have access to it, as prices of food crops soared up while the prices of livestock and charcoal declined. The interagency assessment conducted in the eastern regions of the country recommended immediate interventions in water and food to avert heavy loss of lives.

2.2 Planning and resource mobilisation

The proposal has been developed in a participatory approach in which the staff at the time took part in the planning sessions. The project proposal was submitted to SDC headquarter in December 2003. The programme management team and technical advisors reviewed the proposal and approved. After the comments have been incorporated, it was submitted to DANIDA, which approved it in mid January 2004. Initial preparation such as recruitment and training of the project staff plus mobilization of resources (inputs) took the whole month of February 2004.

Unlike other organizations, which have rigid procedures, long processes and hierarchy, SCD and DANIDA have reacted positively and responsibly to the emergency project, which considerably contributed to relevance and effectiveness of the project. It has also met the standard of emergency fund mobilizations as it took less than a month to be approved in both organisations.

2.3 Hierarchy of the project objectives and outputs

Goal

The goal of the project written as project impact is to contribute to the improvement of the food security situation of the affected people in the Togdheer region of Somaliland.

Purpose

Purpose of the project written as project objective is to create and improved situation for survival, development and protection of 30,000 people of which 6000 are children within in the targeted villages of Togdheer region of Somaliland.

Outputs of the project
- Access to basic food items provided for 30,000 affected people including 6000 children for 3 months from March to May 2004.
- Access to educational materials support provided to 1500 children selected from 15 schools of Togdheer region.
- Strengthened capacity of the disaster committee including community committees through training.

Activities

Conduct community mobilization
- Conduct rapid assessment and identify target beneficiaries
- Identify distribution centres.
- Allocate beneficiaries to identified distribution centres.
- Purchase relief inputs
- Transport relief inputs.
- Ensure safety and security of relief inputs.
- Distribution of the relief food to identified beneficiaries
- Provide educational support to needy children
- Conduct early warning training.
- Conduct training on the special needs of children in times of drought
- Conduct training on targeting of vulnerable groups.
- Conduct networking with UN and INGOs.
- Conduct regular monitoring and review.
- Conduct regular reviews
- Undertake evaluation of the project.

2.4 Project inputs and budget

From the documents reviewed and discussions with the concerned persons, the project input in terms of food grain is 914.53 MT (745 MT 0f rice, 124.2 MT of sugar and 45.75 MT vegetable oils. The monetary value of these food commodities is USD 349648.79. In addition USD 55454.54 in cash had been allocated to implement the project.

3. EVALUATION

3.1. Purpose of the evaluation

The purpose of the evaluation is to provide information on the impact, food economy, and appropriateness of the intervention to draw lessons and make recommendations for future humanitarian response.

3.2. Evaluation process and methodology applied

The Evaluation involved all stakeholders at different levels including those from Hargeisa, Burao of Togdheer region, and community levels. Individuals and agencies involved were contacted. Communities were interviewed. Views of some possible key informants were taken. Thus the evaluation is carried out in a participatory manner, applying the following methodologies for data and information collection:

Document review: Relevant documents including project proposals, assessment reports, situation reports, survey reports, agreements, sphere handbook and tender documents have been reviewed/referred. Field visit and semi-structure interviews: Six communities have been visited and with a pre-prepared checklist, Village committee members, Women groups as well as community and beneficiary groups, in the visited 6 communities have been interviewed. Refer to annex: 3 for the content of the checklist.

Discussions with agencies and key informants: Discussions have been undertaken with the SCD staff in Hargeisa, project field staff in Togdheer region, WFP, HOVOYOCO, CLHE, WFP, OXFAM GB, SC-UK, NERAD, RDC and RG etc. For detailed list see Annex 1: list of organizations and persons contacted.

3.3. Strengths and constraints of the evaluation

The participatory approach applied enabled the evaluation to incorporate information from various sources. As much as possible information was triangulated or crosschecked. Views and opinions of various stakeholders were considered. This was possible because of the cooperation and collaboration of stakeholders in general and the SCD staff in particular.

The worked done include both evaluation and analysis of household economy of rural target pastoral communities in the region. Despite the fact that additional work was done, the duration of the whole exercise was only 15 effective days. See annex 2 terms of reference for the break down of working days. Only 6 of 41 target villages have been visited. There could be a question of sampling size although the information collected from the 6 villages was more or less similar. In each of the visited community, 3 households have been interviewed to collect more information on household food economy. It was not possible to conduct general staff meeting to collect as information as possible from the staff as they were no longer working with SCD.

4. MAJOR FINDINGS

4.1. Project designing and planning

4.1.1. Assessments and surveys

As the drought situation has been getting worse from time to time since 2002, interagency assessment has been conducted in December 2003 and later nutritional surveys at different times. The assessment findings and recommendations were the basis of proposal development and geographical coverage of the target areas. The choice of free food distribution has been also agreed in the inter agency coordination meetings where SCD was appointed as food distribution focal point for INGOs. Adequate assessments and nutritional assessments to decide necessary interventions have been carried out.

4.2. Project pre-implementation activities

4.2.1. Staff training

As soon as the staff were hired and inducted, the training on targeting beneficiaries has been conducted by a consultant to enable the staff to mobilize the community effectively and jointly conduct targeting. The training took 2 days from 3-4 March 2004. Around 7 staff and 2 members from regional office attended the training on targeting. However, no members from the target communities have been invited to attend which could facilitate the targeting process and effectiveness of the targeting.

4.2.2. Community mobilization

The trained staff on targeting mobilized the target communities, started initial assessment and targeting process and at the same time. It was the first time that the staff met the target communities as these target communities have been selected based on the inter-agency assessment which none of them might have participated. The interagency assessment estimated the population of the target communities and provided rough figures. The project based the resource allocation on those figures. As there was almost three months difference between the assessment and community mobilization, the population figures have changed considerably. In some areas it increased while it decreased in others as pastoralists have moved in search of pasture and water. The communities have been informed or given fixed quota (the number of the beneficiaries) to be selected as this had been done at desk review. In addition to inter agency assessment another assessment should have been done before community mobilizations and should have provided adequate and relevant information to the planning process instead of desk planning. This could have also improved community participation in the planning process. Although variations of population figures have been reported, it is questionable if at all they have been incorporated into the planning of resource allocation. This has negatively affected the targeting and registration of the vulnerable households.

Community mobilization generally was fairly satisfactory given the short duration of the project. The target communities were represented through elected community representatives (village committees) for targeting and distribution purpose. However, the some of the committees were found to be less effective than others. This is attributed to insufficient mobilization and absence of a continuous monitoring system of the performance of the committee members through public meeting etc.

4.3. Project implementation

4.3.1. Registration

The registration was meant primarily to identify eligible beneficiaries based on the degree of vulnerability and emergency food needs. It was also intended to get general information about the total population and ideas on the average household size. The generally accepted household size is 6 members although discussions among the INGOs and UN-agencies at regional level have been on made on this point. There is an ongoing process of changing the household figure to 8 instead of 6.

The SCD project is the first organization that initiated the registration of the vulnerable households. Due to the drought situations, most of the villages have been deserted at the beginning. The project registered the vulnerable households left behind as the first point of targeting the needy households. It also provided identification cards to all those registered for the intervention. All those registered received the full ration allocated to them especially at the first distribution. The project was to some extent flexible and adaptable to the changing nature of the affected pastoral communities in that it registered some out migrated households in other villages and delivered food at villages where the households moved in search of better water and pasture.

However, though successful at the beginning, registration later differentiated the households at the same level of need and vulnerability. This is because the project has not succeeded in fully and continuously updating the registered households as the other needy households have returned to their original settlements. This was the biggest challenge facing registration of affected pastoral households. Some of the project stakeholders especially the regional authority and regional drought committee have questioned validity of registering pastoral households who move here and there depending on the rainfall and pasture availability arguing that the registered households will not receive food if they migrate to other areas. The registration of households was found useful and necessary. Most vulnerable households will be identified wherever they are and will be provided food.

Registration has reduced to the extent to which unintended beneficiary could receive food. It also reduced food looting and robbery as some of the communities mentioned happening of such events where there was no registration in the WFP programme. It will also prevent households to receive food from two or more distribution sites. The only point to consider is to avoid having a fixed quota in the registration, continuously update the list of registered households. Households in the out-migrated area will decrease while that of in migrated villages will increase. This will demand or necessitate continuous monitoring of the performance and effectiveness of the project. Delivering food to the households where they are can also reduce registration errors.

4.3.2. Beneficiary targeting

According to sphere standards of targeting, the purpose of targeting is to meet the needs of the most vulnerable while providing aid efficiently and in a way that minimises dependency. The process of emergency project implementation is interdependent. The error in one stage can affect the other or next stage. Targeting is closely linked with the registration so does distribution to targeting. Problem in registration has to some degree affected the targeting of beneficiaries. Those who missed the registration for one reason or another have not obviously been targeted, though measures had been taken during the second and third distributions.

Targeting seemed to have worked well in the first distribution. The most important success is that the most vulnerable groups have been given priority over the others. All of them if not absent received and collected the allocated full rations though shared with others. Project targeting was relatively much better than that of WFP which community called `sit on the sack' meaning every person in the village even the travellers could get food if they meet the distribution by chance. The beneficiary was required to be present during the distribution to be eligible to receive food. No food for the weak and vulnerable groups unable to attend the distribution. According to the communities, if two persons were to share the food, the stronger will take all of the food if they don't know each other. Strong members of same households could receive food from WFP at the expense of the vulnerable and weak households.

Even though the targeting has been generally accepted and considered as fair, awareness and community mobilization both among the staff, and community was low in the target areas. Targeting of beneficiaries has not been standardized as well. Different communities have done targeting in differently on various criteria and varied support. In some of the communities, only village committee has done, others were done by a committee composed of both men and women and still only women have done the targeting in Balli-hiile after the men completely did not agree with project staff how that the targeting was to be done. The common criteria include number of livestock owned, condition of the animals especially whether the livestock could be sold, availability of other sources of income like remittance, physical and mental disabilities, pregnancy and lactation and the wealth group (level of poorness). Various information and observations also indicate that there is error of inclusion. It seemed that objective of targeting was not clear to the communities.

Participation of targeting has not been done according to the sphere standards. Participation of women in most of the area was limited. It was only reported the women did the targeting in Balli-hiile after the complete disagreement among the project staff and community committees or elders. It was rarely reported that the targeting criteria have been well disseminated at a public gathering before actual targeting and registration to avoid suspicion and confusion. The support of the project staff in targeting seemed to have been limited. Some of the communities have been left to do it themselves after being given the quota or the number of the vulnerable households, which negatively affected the targeting itself. For example, Warimran community was informed to target only 70 households based on the nutritional screening. Although the community accepted the offer, yet they actually targeted another 70 although a woman committee member reported that the number was increased to 200 including the agreed 70 households. In Balli-dhiig, the target largely failed and was done on clan system instead. In principle, if 90% of the population is assisted, there is no need to spend time, energy and money to target beneficiaries. Blanket distribution would have been appropriate than going through the targeting process to exclude the remaining 10% of the population. The short-term duration of project in general and inadequate time for community awareness-raising as well as critical drought situation at the time can be the reasons for ineffective targeting.

Generally community targeting is a process and it goes with high level of awareness on the purpose of relief food and the effect of over aid. In traditional and tribal based communities it is difficult for VCs to come out courageously and identify people only on need basis. Therefore it requires longer-term effort to change the attitudes of the people through awareness creation programme.

4.3.3. Distribution

After the food was delivered by transporters at temporary distribution sites and received by the staff, the staff in the presence of village committees commenced distribution of the food and ensured that the already registered households received the allocated quality and quantity of the food. For the first distribution, all registered households received the food ration of one month. While the second distribution consisted of two months with only 25 kg of extra added to the already planned food ration because of the inflation of food commodities. SCD distributed food was highly appropriate and acceptable to the local conditions. The project distributed rice, sugar and vegetable oils, which are the food basket of the region. It was also safe and was fit to the human consumption. The food handling, supply chain management, transportation was up to the standard.

The first distribution of one-month ration was made in March as planned. The 2nd and 3rd distributions were made at one go in the last week of May and first week of June 2004 because of the start of the rainy season. The final distribution took place in mid October. Proximity of the distribution sites to the villages has made the beneficiaries happy. The announcement of the distribution time well before the actual distribution is consistent to Sphere Standards. Most of the recipients of the food were women. The project ensured that the food reached the family and avoided that irresponsible men misuse the food. If the women were absent for any reason, then the distributing staff and the community ensured that responsible men or husbands took the food to the household. Female-headed households were considered seriously so that they will not be excluded. The community generally accepted the principle and understood its value.

In every village, a temporary distribution site has been established. Maximum effort had been made to bring the food near to the beneficiaries. The number of distribution sites is equal to the number of target villages. This has saved the beneficiaries from paying for transport, as they lost most their pack camels. This means that the food went directly to the households. Every household was informed about the size of the food it will receive. The role of each actor, the procedure, distribution time and other related pieces of information were communicated to the community in general and to the VCs in particular. When the food arrived, the VCs were informed well in advance before the distribution day. This information helped them to claim the food they are eligible to. The distributions were carried out openly in the presence of the people. The food was received and distributed by the staff. Every beneficiary physically collected the food. In general terms, the distribution was also fair and responsive, transparent, and appropriate to local conditions. In fact, the SCD food distribution project was the best compared with other projects in the country in general and the affected areas in particular because of provision of locally appropriate food with relatively higher quantity as well as timeliness of food distribution. The communities have the maximum respect for SCD and would rank it as first closest organization if PRA institution analysis were conducted in the region.

However, the quantity of food provided consisting of 50 kg of rice, 10 kg of sugar and 3.65 kg of vegetable oils per 6 member households per month was close to the Sphere minimum standards though deficient in proteins and other essential micronutrients. The food could be adequate for 3 weeks instead of one month. Protein was very essential, as the communities have had no other sources of protein as animals could produce meat or milk. The distribution also timely except the fact that the WFP was reported to have delayed the food distribution on their turn because WFP and SCD agreed to distribute food in the target villages alternatively. The participation of women VC members was limited.

Some other irregularities in distributions have also been observed. Names of registered people who were not present during the distributions have not received the food. Instead, unregistered needy households have been provided on spot with the decision of village committees. The project staff did not want to store the food under the control of the village committee members for later distribution fearing that it might be misused.

4.3.4. Gender aspect of the project

Gender relation in Somali community is constrained by cultural factors. Men and women do not sit together in public meetings in some areas. Relatively there is better environment in urban centres than rural areas. With all the cultural barriers, the project has attempted to involve both men and women in all activities. With the efforts made, it was managed to include women in the VCs .Out of the VC committee members who varied in number across the target areas, 25% have been women. 2 female monitors were recruited mainly to enable SCD to have better access and relation with women beneficiaries. In Balli-hiile, the women reported that they had done targeting of vulnerable households after men have failed to do so.

Regarding the benefit, the project ensured registration of women in their names for their households. Relatively fewer men have also been registered if there were no women in the households. The project has succeeded and enabled that majority of women to receive food ration. Men who lost their wives or whose wives were not able to attend for one reason or the other have received food on their behalf. However, the participation of women as members of the committees was very low. There has a gender conflict in Balli-dhiig. This was caused by misinformation from the project staff claiming that the organisation (SCD) would only contact and deal with women in targeting, food distribution and management. This encouraged women to have 3 days of consecutive meeting discussing the appointment of women committee representing the village. They also generated income for their movement. Such information was reportedly widespread. Men felt that their responsibilities have been deprived of. The men threatened to force all organizations not to enter the town. As part of the revenge, one of the committee in the village took a gun and told the evaluation team to leave the village if they are from any organization. The committee settled the issue by the persuading the gunman and finally the team was allowed to have discussions with the community. Gender issues should be carefully handled as this can lead to unintended result that can damage the gender balance, as it would sideline men. Instead men should be encouraged to give an equal chances in decision-making responsibility with women.

4.4 Project monitoring

The project is expected to monitor its effectiveness in responding to problems identified and addressing the needs of the community as well as changes in broader context of the project continuously to improve the project and achieve the objectives set. Effective monitoring of project activities and post distribution monitoring of the project was inadequate. The population of the target areas have not been closely and continuously monitored.

There was a fluctuation of the beneficiary figures, which seriously affected the registration, targeting, and distribution of the food aid. Monitoring was limited to monitoring of activities and project inputs. Monitoring of indicators was not systematically carried out except information collected during targeting and distribution. After distribution, the staff have been terminated. They could have been useful in monitoring effectiveness of distribution, analysing vulnerability, updating the registration list and monitoring the already set indicators. The senior management support to monitoring of the effectiveness of the project and change in the context was greatly hampered by the security incident and subsequent evacuation. During project implementation, a Kenyan lady working for GTZ has been killed on the route between Hargeisa, the capital and Burao, the regional capital of the target areas. The culprits were captured in Dhoqoshey, one of the target villages on the same day.

4.5 Resource management and utilization (efficiency of the project)

The SCD project efficiently used human, material and financial resources at its disposal. Human resources and transport have been hired during targeting and distribution and terminated as soon as the workload declined. All the financial resources have fully been utilized to address the needs of the most affected communities. The project distributed 745MT of rice, 124.2 MT of sugar and 45.75 MT of vegetable oil to 5000 HHs. The plan was to distribute 750 MT of rice, 150 MT of sugar and 54 MT of vegetable oil. The planned quantity was reduced due to inflation of food commodity prices.

SCD carried out regular and timely budget monitoring whereby surplus budget amount was used to make additional food purchase and distribution to 22 communities, after the approval of no cost extension by DANIDA. The planned costs of the project are almost similar to the actual cost at the end of the project period.

4.6 Appropriateness (relevance) of the project

All assessments and nutritional surveys conducted suggested that emergency situation was prevailing. Even various stress indicators were observed. Migration, shortage of food and water, distress, sale of assets, incomparable TOT, deterioration of health, increased malnutrition were reported in all the assessments and surveys. The measurement for emergency should not be only the number of people that die, but the degree of human suffering too.

When asked the situation before the intervention, the community verified the information in the assessment and nutritional surveys. The community informed that most urgent needs of the community were access to food and water among the other environmental degradations and deteriorating TOT. This is because the livestock could not meet the needs of the communities as their conditions were deteriorating from time to time due to lack of pasture and water.

The project was highly relevant to the needs of the community as it addressed priority needs of the community (food). The communities have been also consulted on the food choices and food basket. In line with this, the project provided rice, sugar and vegetable oils, the common food basket in the region. The community also mentioned that the project inputs such as distribution of food started at a time when they had exhausted their livelihood base and were on verge of starvation. Apart from the sharing with needy relatives, there was no exchange of food or sale of food reported in any of the communities. When asked the reason, they mentioned that what they needed was only these and there was nothing better to exchange with the required food.

4.7 Coverage of the project

The assessment and nutritional surveys recommended the most severely affected areas in the region. It also estimated that around 90,000 people were in need of immediate intervention to avert the full scale of starvation and death of the affected communities. SCD targeted 30,000 of 90,000 affected communities in the region especially the villages in the south and south east of the regional capital.

Although the project covered the needs of some of the most vulnerable groups within the community, the needs were high as compared to project input. This is because the situation seemed to have been worsening from time to time and many people considered as better off initially slipped down to the most vulnerable category to the extent that there was no difference between the two groups. This has negatively affected targeting and distribution of food aid. As the performance and effectiveness of the project was not closely monitored, the project did not fully succeed in updating the list of beneficiaries and their situation most of the time. In every community visited reported that small portion of the population benefited from the project inputs as compared to the population. The community themselves had tried to solve the problems by sharing with those who did not receive any of the food. These were given half of what of the registered households received or sometimes below that. This was particularly very clear in the second distribution when the project had to provide full ration for two months. The project therefore has not covered the needs of all vulnerable groups effectively.

4.8. Coherence (coordination and collaboration) of the project

As the region was quite large and the communities have been suffering from severe food insecurity, SCD, WFP and two local NGOs (HOVOYOCO and Candlelight) have intervened in the food sector. Weekly and monthly coordination meetings have arranged at national and regional levels. Initially the information was shared in the different level coordination meetings. Generally the target areas have been assigned to the different organizations. HOVOYOCO has been assigned to north and eastern parts of the regions, and Candlelight to the west. SCD and WFP targeted the south and southeast of the region in the same target areas. As this division of geographical areas and sectors of intervention have been agreed in the coordination meetings, the coordination meetings died down at national and regional levels because the local authority has stopped inviting the agencies as representatives from agencies did not show up in a number of meetings. Other key informants mentioned that need for the coordination meetings declined as the geographical areas and sectors of intervention have been clearly defined.

The coordination of project with NGOs and UN-agencies was relatively better than with local authority. It was also satisfactory in general terms. There was no an overlap of activities as such. SCD collaborated with Oxfam GB in capacity building of the National Environmental Research and Disaster Preparedness Authority (NERAD) and SC-UK in implementing educational component of the emergency project. However, the agencies involved in food have used different approaches of targeting, distribution of food. They distributed different quality and quantity of food commodities. For example, LNGOs distributed 25kgs of rice, 10 kg of sugar and 3.65 kg of vegetable oil per 6-member family/month whereas SCD provided 50 kg of rice, 10 kg of sugar and 3.65 kg of vegetable oil per month for the same family size. WFP provided 90 kg of maize, 10-15 kg of pulses and 3.65 kg of vegetable. Last distribution, WFP provided 50kg of poor quality rice, 10-15 kg of pulses and 3.65 kg of oil.

This has made a lot of confusion on the target communities. SCD and WFP having the same target communities have agreed to alternatively distribute food on monthly basis. This was because the WFP requested SCD to do so as they provided low quality maize culturally unacceptable to the community. Both organizations have failed to harmonize their approaches of food distribution, targeting so as complement each other's programme. WFP used top down approach while SCD on the contrary used more of the participatory approach. Alternative distribution from the two organizations caused confusion to the target communities. It would have been better if the two organizations could divide the target villages because of the difference in the amount of food, type of food, quality of food, appropriates and acceptability of food, as well as different approaches of food targeting and distribution.

4.9. Connectedness of the project

The project was initially planned for three months duration with aim of improving food security situation of the drought-affected communities. The project activities have been carried out in sustainable manner although arrangement with WFP extended the duration and sustainability of the project activities was affected. The target communities have gained a lot of experience in food targeting and distribution. Relationships with SCD have formed which could be useful for similar future programme In the project proposal, the short-term emergency response has been designed in a context, which takes the long-term problems into account. One of the outputs was to strengthen the capacity of the disaster committee including community committees through training. Sphere training covering food security, nutrition, food aid and special needs of the vulnerable groups in general and children in particular has been conducted for national and regional authorities. However the participants from the wide geographical target areas were relatively low.

4.10. Project achievement and impact

The project achievement can be described in the activities accomplished and in the achievement of the expected outputs while the impact would be sought in the extent to which the project objectives are met.

4.10.1. Activities planned versus actual achievement

The project has carried out most of the planned. The recruitment of project staff and the training of staff on targeting, mobilisation of the community, identification of beneficiaries, and distribution sites, purchasing and transportation of relief food, monitoring for efficiency have been carried out as planned. Sphere training covering food security, nutrition, food aid and special needs of vulnerable groups in general and children in particular in times of drought have been carried out for security reasons. Nevertheless, training on early warning has not been implemented for security reasons. The strategic plan and national policy on disaster have replaced the training on early warning.

4.10.2. Achievement of the expected outputs

The accomplished activities that could fall under the first and third outputs have led to the achievement of the most important project outputs. Inflation of food commodities affected the project inputs and reduced quantity of planned food from 750 MT to 745 MT of rice, from 150 MT to 124.2 MT of sugar and from 54 MT to 45.75 MT of vegetable oils. This is less than 1% for rice, 17% for sugar and 16% for vegetable oils. On average, 11% of the total food was reduced even when the surplus money obtained from various budget lines have been spent. On average, the project achieved the most important outputs of food distribution of 914.5 MT to the planned beneficiaries of 5000HHs in the target areas.

The third output `strengthened capacity of disaster committee including community committees through training' have also been achieved. Sphere training on food security, nutrition and food aid, as well as special needs of the vulnerable groups has been conducted. NERAD strategic plan and disaster policy have been developed.

However, the second project output `access to educational materials support provided to the 1500 children selected from 15 schools of the region'' implemented through SC-UK is not fully achieved as some of the visited communities have not verified distribution of educational materials. SCUK asserted that most of the educational materials were provided to schools to be used by teachers and students at school level. Schools in three villages visited received the materials consisting of registers, white papers, teaching aids etc that had no direct impact on children's education and motivation. It would have been better if exercise books, pens and pencils provided to the children in addition to support given to schools.

The project nevertheless has succeeded in producing the core project outputs. The core outputs were the most important for the realization of the purpose of the project and contribution to the project goal. 4.10.3. Achievement of project purpose and goal (Project outcome and impact) The achievement of the core project outputs such as `access to basic food items' and strengthened capacity of disaster committee including community through training have led to the realization of the project purpose which was to create improved situation for survival, development and protection of the affected community in Togdheer region.

Having achieved the purpose, the food distribution averted the eminence of large scale of famine by decreasing hunger, starvation and related diseases and ultimately heavy loss of lives. According to the communities, the food distribution started at time when communities exhausted their coping capacities. Many vulnerable groups have been left behind with no food to eat. Many other households were unable to cook even a single meal. In Ballihiile, the community informed that they have, during distribution, provided small amount of food (few kg of rice) to the very vulnerable groups who had nothing to eat to prepare lunch, as they were unable for wait for the few hours of food distribution. It enabled these vulnerable groups to survive. The community in Nasiye informed that the food distribution had reduced plight and worry as well as frustration of food crisis. It also saved livelihood base as selling of remaining few breeding animals have been spared. The communities also mentioned that purchase of food from the market on credit or by any other means declined as appropriate food became available. In Balli-dhiig, the communities mentioned that it reduced begging of food by vulnerable groups as they received food. Warimran community also reported that the food distribution reduced the burden of the community on collecting food contribution to the needy and vulnerable households in the village. All the communities visited were happy with food and grateful to SCD and Danish people.

The project had also a significant impact on the child education. As the migration continued in the most of the villages, many children dropped out, as they had to move with their families. Because of this reason, the many schools were on the verge of closing down as the student population decreased. In Balli-hiille eleven children have been left at villages with few kg of ration, which could run out immediately. With food distribution, the children were able to continue education as they had received their share of the food. They themselves started to cook. The project had a significant impact on continuation of child education. In addition, it contributed to the improvement of nutrition and reduced deteriorating situation of malnutrition.

The food distribution also reduced further migration either to towns or to other villages in search of food, pasture and water. The project additionally brought about social benefits to the recipient households in reducing divorce of the existing families because of lack of food and income to satisfy their needs and increased marriage. The community in Balli-hiille reported the marriage increased in the village after the food distribution. At the time of evaluation, the team observed preparation of a wedding ceremony.

Of the only negative impacts that the project brought about is the gender conflict in Balli-dhiig. Sign of dependency syndrome has also been observed, as people will ask the strangers what they brought to their villages.

5. CONCLUSIONS AND RECOMMENDATIONS

5.1 Although the project covered the needs of some of the most vulnerable groups within the target communities satisfactorily, the needs were high in the target areas as compared to project inputs. This is because the situation seemed to have been worsening from time to time and many people considered as better off initially slipped down to the most vulnerable category to the extent that there was no difference between the two groups. This has negatively affected effectiveness of the targeting, registration and distribution of the food as well as project coverage. The project needs to improve its coverage.

5.2 The project was very relevant to the needs of the community as it addressed the priority need of community (food). The communities have been also consulted on the food choices and food basket. In line with this, the project provided rice, sugar and vegetable oils, the common food basket in the region. The community also mentioned that the project inputs such as distribution of food started at a time when they had exhausted their livelihood base and were on verge of starvation. Apart from the sharing with needy relatives, there was no exchange of food or sale of food reported in any of the communities. When asked the reason, they mentioned that what they needed was only these and there was nothing better to exchange with the required food.

5.3 The SCD project efficiently used human, material and financial resources at its disposal. Human resources and transport have been hired during targeting and distribution and terminated as soon as the workload declined. All the financial resources have fully been utilized to address the needs of the most affected communities. The project distributed 745MT of rice, 124.2 MT of sugar and 45.75 MT of vegetable oil to 5000HHs as compared to the planned 750 MT of rice, 150 MT of sugar and 54 MT of vegetable oil.

5.3 Despite the fact that other agencies did not agree on overall approach of dealing with community, food targeting, distribution, type of food, quantity and quality of food to be distributed in the target areas because of their different approaches, SCD's project coordination and collaboration with other agencies and local authorities was satisfactory. Weekly and monthly coordination meetings have been arranged at national and regional levels at the onset of interventions. Geographical target locations and sectors of intervention have been divided. Roles and responsibilities of different organizations were clearly defined. Information on progress and constraints were shared in the different level coordination meetings thought coordination meetings later became nominal.

5.4 The project has carried out most of the planned activities. The accomplished activities that could fall under the first and third outputs have led to the achievement of the most important project outputs. Inflation of food commodities affected the project inputs and reduced quantity of planned food from 750 MT to 745 MT of rice, from 150 MT to 124.2 MT of sugar and from 54 MT to 45.75 MT of vegetable oils. This is less than 1% for rice, 17% for sugar and 16% for vegetable oils. On average, 11% of the total food was reduced even when the surplus money obtained from various budget lines have been spent. On average, the project achieved the most important outputs of food distribution of 914.5 MT to the planned beneficiaries of 5000HHs in the target areas. The third output `strengthened capacity of disaster committee including community committees through training' have also been achieved. Sphere training on food security, nutrition and food aid, as well as special needs of the vulnerable groups has been conducted. NERAD strategic plan and disaster policy have been developed.

5.5 Having achieved the purpose of the project, the food distribution averted the eminence of large scale of famine by decreasing hunger, starvation and related diseases and ultimately heavy loss of lives. It enabled the target group to survive and reduced worry and frustration of food crisis. It also saved livelihood base as selling of remaining few breeding animals have been spared. The communities also mentioned that purchase of food from the market on credit or by any other means declined as appropriate food became available. It has also reduced begging of food by vulnerable groups as they received food and decreased the burden of the community on collecting food contribution to the needy and vulnerable households. The project had also a significant impact on the child education. It enabled continuation of child education in the drought-affected community. School children and their families received their share of the food.

Furthermore, the food distribution also reduced further migration either to towns or to other villages in search of food, pasture and water. The project additionally brought about social benefits to the recipient households and reduced divorce of the existing families because of lack of food and income to satisfy their needs and increased marriage. In addition, it contributed to the improvement of nutrition and reduced deteriorating situation of malnutrition.

5.6 Of the only negative impacts that the project brought about is the gender conflict in Balli-dhiig. Signs of dependency syndrome has been felt and observed, as people will ask you what you brought to their villages.

5.7 From Sphere Standards perspective, the project met most of the food aid standards such as response, food appropriateness and acceptability, food safety, food handling, supply chain management and was close to initial assessment, ration planning, food quality and distribution. However it needs to improve the common standards of participation, targeting, monitoring and aid workers competencies and responsibilities.

Having made these conclusions, the consultants would like to forward the following recommendations and suggestions.

5.8 Depending on difference in time between the inter-agency assessment and actual start of the project, additional initial assessment should be conducted to validate and cross-check population figures, change of needs, and changing overall context of the emergency situation and should provide the sufficient and relevant information for planning. Planning should not be done at desk and should be based on the updated initial assessment.

5.9 Community should be fully mobilized to create general awareness of the intervention. SCD should facilitate selection of committee, which will manage the food distribution, carry out targeting and registration as well as any other necessary activities. Electing the village committee will be advantageous in ensuring gender balance, participation of different categories of the community and carrying out the work delegated to them effectively. The community should be told that if the committees fail to do their responsibility that they will be replaced. If the communities accepted the already existing community, additional members from other categories would have been encouraged to join.

5.10 The committee should be trained on targeting, registration and food distribution and management to facilitate their functions. The staff and committee should have a clear concept of the targeting criteria and targeting mechanism. These criteria should be also approved at the public meeting so that individuals and households will be clear why or why not if they are not targeted. Retargeting should be done if there are changes in the vulnerabilities and list should be updated. This should be also approved at public meetings as well.

5.11 Targeting mechanisms and process should be standardized across the target areas. The staff and elected community committee should do the targeting uniformly with clear objective and understanding. The targeting process should be part of the training on targeting and should be followed by all staff and committees seriously.

5.12 Fixed quota for individual communities should be done after the organizational initial assessments that they have participated the communities. If criteria are accepted at public meetings, the process of targeting and registration of household will be straightforward. The quota can be communicated to the communities. The RPC with support of the PM will allocate resources based on the assessment findings and update accordingly after every monitoring of the situation.

5.13 Somali culture is man dominated one. Change of gender relation needs enough time as it is involved in attitudinal, behavioral, social and cultural change. Gender issues should be handled carefully to avoid gender conflict. The staff should have a full understanding of the gender issues and refrain from sharing sensitive gender information that can have negative impact on SCD work in the drought affected communities. Gender should not be interpreted as women issues but the relationships between men and women in the society. Possibility of gender training can't be ruled out.

5.14 Monitoring system capturing the key issues in the project should be put in place before implementation is initiated. Both monitoring for efficiency and effectiveness should be carried out in the project lifespan. The monitoring system should be analyzed and used in the project to improve performance and effectiveness. Key indicators at output, purpose and goal levels should be monitored to understand if the project is achieving the already set objectives. Adequate attention should be given to post distribution monitoring as well. Role and responsibilities of different actors in the project monitoring should be clarified in the monitoring system. SCD is recommended to hire a consultant on establishment of effective monitoring system of the project who will train staff and relevant community members and local authorities.

5.15 The project should increase the level of community participation. Decision-making process of food targeting, registration and distribution should be in the hands of the community. On arrival, the community should receive and take over the responsibility of the food and do distribution by themselves with support and supervision of the project staff. A degree of trust with the community should be established and the communities take minimum decisions.

5.16 Of the recommendation include provision of 75kg of rice, 20kg of dried beans, 7.2 kg of vegetable oil/household/month. This will meet the sphere minimum standards in terms of energy, protein and fats though still deficient in some other essential micronutrients, which the target communities can gain from other sources. The project should advocate for meeting of Sphere standards.

5.17 The project is recommended to increase the coverage of vulnerable groups as most of them lost their livelihood base. The number of targeted beneficiaries is to be increased to 7500-10,000 households depending on resource availability. The project should have separate target areas to avoid community confusion on the approach of food targeting and distribution, type of food, quantity and quality of food as well as timing of distribution. It will improve capacity building of the target communities, as they will focus on SCD approach of working with the community.

5.18 Coordination and collaboration with WFP should not be at expense of the project effectiveness. Impact of the project cannot be effectively monitored or traced if the different organizations alternatively carry out distribution in the same target areas. The target areas should be divided into two so as to complement each other's project and avoid community confusion. The project should encourage WFP to provide appropriate and acceptable food items with the right quality and quantity.

5.19 The project should continue the free food distribution up to June 2005 as the dry period (Jilal) is to start where food security situation in drought affected pastoral community is not expected to improve in the near future. Technical feasibility of food for work component for households who are able to work is to be studied to reduce dependency syndrome. This will have double benefits. It will first enable the affected population to acquire the food and secondly rehabilitate the rangelands and environment, which are the basic causes of the recurrent droughts in the country. If it is not technically feasible, possibility of working with partners can be considered to do this piece of work

5.20 The project is recommended to formulate an exit strategy and design in the project proposal. SCD should link up with those organizations, which are planning drought recovery programmes such as Oxfam GB, PENHA, HOVOYOCO, CLHE, and sort out the issue before the next rainy season in May 2005.