World Food Programme: Fighting Hunger Worldwide WFP Yemen Crisis Response Pre-assistance Baseline Survey Highlights This baseline survey was conducted in June 2015 in seven governorates (Aden, Al Hudaydah, Amran, Hajjah, Lahj, Sana a City, and Taizz) to set a baseline of WFP s emergency operation in Yemen after the escalation of the conflict in March 2015. The interviewed households were planned beneficiary households under WFP s emergency operation, thus the findings show the pre-assistance food security situation of the most vulnerable population and do not represent the entire Yemeni population. The food security situation is alarming with 91% of the planned emergency operation beneficiaries being food insecure. Dietary diversity is limited with three food groups consumed in a week on average most commonly consisting of cereals, oil and vegetable. The three most common consumption-based coping strategies when faced with a shortage of food are to: 1) rely on less preferred and less expensive food (5 days per week); 2) limit portion size of meals (2.8 days per week); and 3) borrow food or rely on help from a relative or friend (2.8 days per week). The four most common livelihood-based coping strategies when faced with a shortage of food are to: 1) purchase food on credit (80% of planned beneficiaries); 2) borrow money (59%); 3) sell household assets/goods (37%); and 4) reduce health and education expenditures (37%).
Background The conflict in Yemen has worsened since late March 2015, posing new challenges for the people of Yemen and the international humanitarian community. Out of the country s 22 governorates, 20 have been affected by the escalated violence, and the latest Integrated Food Security Phase Classification (IPC) findings show that 12.9 million Yemenis are food insecure of which 6.1 million people are in need of urgent food assistance. WFP Yemen has begun an emergency response in April 2015 with plans to gradually scale up its assistance to reach all affected governorates over the coming months. WFP Yemen has conducted a pre-assistance baseline survey to ensure rigorous outcome monitoring and to set baseline values against which the impact of the emergency interventions will be measured. A post-distribution monitoring (PDM) outcome-level household survey will be conducted later in 2015. The PDM will use the same food security indicators as in this pre-assistance baseline survey in order to ensure comparability of beneficiaries food security situation before and after WFP assistance with continuous monitoring of emergency interventions over time. Methodology During the survey, 480 households were interviewed in June before Ramadan in the following seven governorates: Aden, Al Hudaydah, Amran, Hajjah, Lahj, Sana a City, and Taizz 1. A two-stage cluster sampling approach was adopted with 90% confidence level and 5% precision level to have a representative result at the country level (all seven governorates combined). However, the selection of 30 food distribution points (clusters) was not done in a completely random manner. The list of all food distribution points to be targeted under the emergency operation is not yet available, as the operation is being gradually scaled up. Therefore, any food distribution points in the targeted food insecure and conflict affected districts where emergency food assistance had not been distributed since the escalation of the conflict were counted as a cluster in the sampling framework of this baseline survey. Those governorates with a high intervention priority and a high number of beneficiaries were considered as a priority for the baseline survey as well. Coupled with this, considering limits to site accessibility due to continuing airstrikes and street fighting in some areas, WFP Yemen gave priority to governorates and districts with enumerators already living on-site. Given the limited mobility in the country and the importance of local knowledge in negotiating access to conduct household surveys, it was crucial to employ enumerators already from and familiar with the areas in question. Five male enumerators and three female enumerators were employed in total. WFP chose five food distribution points to be surveyed in each governorate as randomly as possible 2. In each food distribution point, 16 households were randomly selected to be interviewed, provided that they met the two following conditions: 1) They have not received any WFP food assistance since the start of the emergency operation in April 2015. 2) They meet one or more of the following WFP targeting criteria for the emergency operation: Displaced households; Households headed by persons unable to earn income from own work such as those headed by widowed, divorced or separated women, elderly and disabled; Poor households from the socially and economically marginalized communities (the muhamasheen); Households that depended on the Social Welfare Fund or community charity as their main source of income; or Households that depended on income earned from daily wage for livelihood and access to food and that no longer have access to casual work. The data entry and analysis was done by the Country Office Monitoring and Evaluation (M&E) Unit using SPSS Software. The findings of this baseline survey is compared with the findings of the same seven governorates of the last 1 The initial plan also included Sa ada governorate. However, due to several logistical and security constraints and the time limit before the start of the month of Ramadan, the survey could not be conducted in Sa ada. 2 Except in Aden (three clusters) and Lahj (two clusters) due to logistical and security constraints. 1
Comprehensive Food Security Survey (CFSS) for which data collection was done in March and April 2014. The CFSS findings are presented as a reference, not for comparison purposes, as the CFSS findings represent average Yemenis while the baseline survey covers the planned emergency operation beneficiaries, which is a more vulnerable subset of the population. Demographics Graph 1 shows how many of the planned beneficiaries belong to each of the WFP targeting criteria. Note that responses were not mutually exclusive so it is possible that respondents may meet several criteria. It is found that 38% of the planned beneficiaries in the surveyed areas depended on income earned from daily wages for their livelihood and access to food before the crisis and no longer have access to casual work. 29% of the planned beneficiaries in the surveyed areas are internally displaced persons (IDPs). Graph 1: Category breakdown of the planned emergency operation beneficiaries About 75% of the planned emergency operation beneficiary households are found to be male-headed, which is in line with previous monitoring exercises among beneficiary households, while the 2014 CFSS found that 91% of the households in the seven surveyed governorates are male-headed. The average household size is seven, which is the same as the national average recognized by the government of Yemen and the humanitarian community. The demographic composition of the households is 48.1% men and 51.9% women, which is very close to the national average of 49.6% men and 50.4% women according to the 2004 Yemen Population Census. Food Consumption The household Food Consumption Score (FCS) module is used as a proxy indicator for household food security. The Food Consumption Score is a measure of dietary diversity, food frequency and the relative nutritional importance of the food consumed. The FCS is used to classify households into three groups: poor, borderline or acceptable food consumption, and the households with acceptable food consumption are considered food secure, while those with borderline or poor food consumption are considered moderately or severely food insecure respectively. The results of the baseline survey indicate an alarming food insecurity level in the seven surveyed governorates in the country. As shown in Graph 2, only 9% of the planned emergency operation beneficiary households are food secure, or able to meet their essential food needs. 29% of them are moderately food insecure with significant food consumption gaps, or are marginally able to meet minimum food needs. The remaining 62% of severely food insecure planned beneficiaries has extreme food consumption gaps. Graph 2: Food consumption level The results of the baseline survey are even more worrisome with the 2014 CFSS as a reference. The level of nationwide food insecurity in the pre-crisis period stood at 39% with 18% of severely food insecure in the same seven surveyed governorates. 2
Dietary Diversity The findings of the baseline survey indicate that dietary diversity for the planned emergency operation beneficiaries is worse than that of the average Yemeni during the pre-crisis period. Graph 3: Dietary Diversity Score (average number of food groups consumed per week) As shown in Graph 3, the planned emergency operation beneficiary households are found to consume three food types per week on average, while average Yemenis during the pre-crisis period consumed five food types per week on average. The three most consumed food types are found to be cereals, oil and vegetable for the planned emergency operation beneficiaries, while the five most consumed food types before were cereals, oil, vegetable, meat, and dairy. Graph 4 shows the stark difference in the share of the households consuming each food group in a week except for cereals and oil. Graph 4: Share of households consuming each food group in a week Consumption-based Coping Strategies The reduced Coping Strategy Index (rcsi) is an additional proxy indicator of household food security to better understand the frequency and severity of changes in food consumption behaviors when faced with a shortage of food. Households were asked how many of the preceding seven days they did not have enough food or money to buy food and adopted one of the coping strategies. The higher the CSI value, the higher the degree of food insecurity. The minimum possible CSI value is 0, while the maximum is 56. Graph 5: Consumption-based reduced Coping Strategy Index (rcsi) 3
As shown in Graph 5, the baseline survey results indicated a very high consumption-based reduced coping strategy index (rcsi) at 19, which indicates that households have to resort to severe measures to cope with the shortage of food. The three most common coping strategies are found to rely on less preferred and less expensive food (5 days per week), to limit portion size of meals (2.8 days per week), and to borrow food or rely on help from a relative or friend (2.8 days per week). The CSI during the pre-crisis period of the average Yemenis was 8. As displayed in Graph 6, the planned emergency operation beneficiaries are found to engage in all five coping strategies more frequently, compared to the average Yemenis during the pre-crisis period in the seven surveyed governorates. Graph 6: Average number of days each coping strategy is used in a week Livelihood-based Coping Strategies The livelihood-based Coping Strategy Index is measured to better understand longerterm household coping capacities. Household livelihood and economic security is determined by income, expenditures and assets. Understanding the behaviors households engage in to Graph 7: Share of households using each category of coping strategies as their maximum level strategies adapt to recent crises provides insights into the difficulty of their situation, and how likely they will be to meet challenges in the future. Households were asked if anyone in their households had to engage in any of the ten coping strategies because there was not enough food or money to buy food during the past 30 days. One neutral strategy, four stress strategies, three crisis strategies, and three emergency strategies were asked based on the severity of the strategies. The higher the CSI value, the higher the degree of food insecurity. 4
As shown in Graph 8, the four most common livelihood-based coping strategies when faced with a shortage of food are to: 1) purchase food on credit (80% of planned beneficiaries); 2) borrow money (59%); 3) sell household assets/goods (37%); and 4) reduce health and education expenditures (37%). Graph 8: Share of households using each coping strategy 5
Mobile Phone Access In light of possible heightened monitoring access constraints and potential remote monitoring through live calls, interviewed households were asked if they had access to mobile phones. As shown in Graph 9, some 65% of the surveyed households reported having one or more mobile phones in the households. Another 20% are found to have access to mobile phones through neighbors, friends, Sheikhs, or school headmasters. The remaining 15% do not have mobile phone access. Among the 65% of the households with mobile phone access in their households, 81% confirmed that they have mobile phone coverage in their area of residence. However, only 52% confirmed that they could recharge their phones on a regular basis. Other External Assistance Graph 9: Share of households with mobile phone access Some 25% of the surveyed households have reported to be currently receiving some sort of aid support from NGOs, humanitarian organizations or authorities/communal institutions. When asked what kind of assistance they receive (responses were not mutually exclusive), only 4% of the entire surveyed beneficiaries reported receiving food aid. 17% of the all beneficiaries answered that they receive assistance in cash/vouchers. Conclusions Though limited to seven governorates, this baseline survey confirmed an alarming level of food insecurity among the planned beneficiary households under WFP s emergency operation, with 91% of them being food insecure. Considering that the situation may be even worse in inaccessible areas and that only 4% of the surveyed beneficiaries reported receiving food aid from other organizations, the urgency of effective implementation of the emergency operation is paramount. The baseline survey further highlights the potential costs of a worsening crisis as the planned beneficiary households rely on less preferred and less expensive food for five days per week on average, and 80% of the planned beneficiary households were found to purchase food on credit. For further information, please contact WFP Yemen M&E Unit: dongeun.kim@wfp.org mahdi.khalil@wfp.org www.wfp.org 6