1. Joint UNHCR-WFP Impact Evaluations of the Contribution of Food Assistance to Durable Solutions in Protracted Refugee Situations 2011 & 2012

2. Emergency Model ‘Care and maintenance approach’ Minimum standards regarding basic needs Food in kind, fuel, shelter, water, cooking utensils, health care, education Encampment Protracted Model Possibility of ‘self-reliance’ as a refugee Livelihoods beyond relief Changing population profile and social context Inputs of two UN agencies working towards same goals New food assistance tools Recognition of wider range of stakeholders /partners Multi-year strategy/plan Durable solutions: going home, resettlement or integration WFP & UNHCR Joint Commitment TIME 

3. Series Objective & Audience

4. APPROACH 4 countries: Ethiopia, Rwanda, Chad & Bangladesh Mixed method theory- based impact evaluation, field-based Common evaluation framework & key questions Joint WFP-UNHCR evaluation management, following WFP evaluation standards Series findings synthesised

5. TimeFood assistance AssumptionsExpected results Short term: Emergency Response General food distribution Full rations: 2,100 kcal per person per day Registration & food distribution systems functioning; food is taken home & consumed; food basket is sufficient; local partners have sufficient capacity Lives saved; improved food consumption; safety and protection provided. Minimal level of self-reliance. Medium term Food assistance decreases (partial rations) Transition from emergency response; complementary social service interventions are available, e.g. water, sanitation, education, housing, etc.; targeting is effective Improved food basket, improved nutritional status (acute and chronic malnutrition). Increased capacity of beneficiaries to establish livelihoods. Long term: > 5 years, protracted Food assistance decreases (partial rations) Livelihood interventions available; asset building; legal status allows for employment; access to land; local institutions provide beneficial services Refugee self-reliance; local integration; resettlement or repatriation. Simplified Logic Model

6. Methods 6 Quantitative Surveys of variability in indicator values across large number of refugee households: Food Consumption Score, Household Dietary Diversity Score, Coping Strategy Index, asset scores) (Stratified random sample, < 7% error Review of existing literature & data (e.g. reports/studies; nutrition data on malnutrition: GAM, SAM & chronic; previous FCS, HDDS, CSI) Participatory methods to understand the ‘why’ smaller sample of refugees, host population, other actors (focus group sessions with different ages & sexes, semi-structured interviews) Evaluation Matrix Triangulation Observation: transect walks in camps & of food distribution

7. Challenges & How Managed No conventional counterfactual  Limits to attribution of results. Focus on contribution.  Selection of other comparisons, relevant to programme context: - Intra- and inter-camp comparisons (Ethiopia & Rwanda) - Comparison between camps where planned levels of food assistance varied (Chad) - comparison with ‘unofficial’ refugees in makeshift camps who received no food assistance (Bangladesh)  Question = how does the impact vary across a large number of stratified and randomly selected households?  Synthesising similarities across contexts/cultures Incomplete/unreliable databases for past years  Assemble evidence from wide range of sources & triangulate  Document limitations Possible biases in data arising from survey timing (Chad & Rwanda) 7

8. In short-term after arrival: hunger mediation achieved coping strategies improved Longer-term: o Unacceptably high percentage of households are not food secure, especially women-headed households o Chronic malnutrition at or above ‘high severity’ in all 4 contexts o Very limited livelihood opportunities, few assets, frequent negative coping strategies o Food and non-food items treated as income to meet unmet basic needs o Protection (especially SGBV) inadequately addressed Desired evolution to greater self-reliance has not been achieved Results

9. Global Acute Malnutrition rates WHO GAM benchmark 10-14% indicates serious situation

10. Why? Contextual Factors: Funding shortfalls: uneasy fit in emergency- development divide Host government policies Implementation Factors Missed opportunities for synergies with other programmes Poor follow-up to JAM’s and weak Joint Plans of Action Inaccurate household records & infrequent revalidation Insufficiently frequent and poor timing of non-food item re-distribution Inadequate monitoring of food distribution

11. Credible, useful & used – enablers

12. With thanks to all host governments and to DARA and TANGO International as well as colleagues in the WFP and UNHCR Evaluation Offices

13. Reference: Full and summary reports of each evaluation in the series, the Synthesis Report and the Management Responses are available at http://www.wfp. org/about/evalua tion/list?type=27 11&tid_1=All&tid _2=All&tid_3=19 59