WFP CAMBODIA VAM TARGETING PROCESS 1998 VAM Targeting Goals: “The aim of a WFP food-aid-assisted recovery programme is to enable people to restore their livelihoods in order to assure immediate and longer-term food needs. WFP can contribute to the process of transforming insecure, fragile conditions into durable, stable situations.” (From Crisis to Recovery, WFP 1998) To help fulfil this aim, the WFP VAM Unit in Cambodia has been assigned to target areas and populations for recovery and rehabilitation with food aid. Targeting has three basic goals in mind. Strengthen understanding of vulnerability and food insecurity among WFP and Counterpart staff Build stronger links between targeting and food insecurity Support 3 tiered targeting strategy for WFP supported rehabilitation activities: area/community/people Basic Targeting Concepts in Cambodia Food insecurity in Cambodia can be observed in a number of ways: 1. Normal entitlements are regularly not enough and cause chronic shortages of income and food. 2. Transient food insecurity when a natural disaster or man made event occurs that temporarily and precipitously reduces entitlements. 3. Dependence on entitlements which carry unacceptable risk, such as going to insecure or mined areas for livelihood, long term indebtedness for food, and transient migration for employment. Outcome Indicators: Information from the national 1997 Socio-economic Household Survey and the 1997 WFP/MRD rapid rural appraisal have been linked to help establish a set of outcome indicators that are used to define the direct results or defining characteristics of food insecurity. These include anthropometric measurements for children under five years old, total estimated expenditures, % of food to total expenditures, and reliance on high risk coping strategies. Regression and correlation analysis helped to determine factors closely linked with specific outcomes. Targeting: Targeting is a resource planning tool, which allows findings of vulnerability analysis to be combined with operational objectives in order to establish priorities and allocations. There are two major processes at work in targeting: measuring indicators of food insecurity relevant to distinct population groups, and monitoring changes which can change the food security status of people. Vulnerability: The notion of vulnerability is a relatively new concept in food security studies, which emphasizes the combination of risks to food availability and people’s abilities to cope with such risks. By looking at both sides of the vulnerability equation, WFP is able to gain a more in-depth and sustainable knowledge of trends and events which affect populations at risk. Operational Parameters- WFP Cambodia’s targeting methodology is based on the intervention strategy outlined in the CSO. This strategy was refined when the Country Office prepared the Protracted Relief and Rehabilitation Programme draft for the period 1999-2000. The broad goal of the PRRO is: Sustained food security among the chronically hungry poor, simultaneously with the promotion of re-emerging social cohesion and support systems. The objectives - in line with WFP's Mission Statement - will be to: i) build and rehabilitate essential assets, facilitate reconciliation and promote self- reliance among the poor; ii) contribute to improving quality of life; iii) provide immediate relief from hunger among specific groups temporarily unable to meet their basic food needs. The specific outputs for reaching this objective are as follows: Activity Objective 1.1 Rural road improvement Economic growth, market integration & food security 1.2 Rural road maintenance Medium-term sustainment of road improvement 1.3 Pond & well installation benefits Time saved; better health; vegetable & livestock 1.4 Water control infrastructure prodn Increased yields & security of food production 1.5 Land clearance Facilitation of resettlement & reconciliation 1.6 Village wood lots Fuel security; time saved & environmental 1.7 Rice/seed banks conservation Credit in support of rice production & food security 1.8 Latrine construction Reduced diarrhoea incidence & general improved 1.9 Cultural heritage conservation health Development of the tourism industry General Steps taken by the VAM Unit for 1998 Targeting 1) Establish typology of food security zones in Cambodia, defined by land use, entitlements and physical features. Assign zones by district in Cambodia 2) Identify outcome indicators to food insecurity 3) Through analysis of VAM database, identify determinants and proxy indicators available at district level for identifying vulnerable districts. 4) Organise results and review findings with field staff and Ministry of Rural Development counterparts. 5) Through continued analysis of household information and participatory methods, assist the Country Office to target most vulnerable households. Food Security Zones- In previous targeting exercises, WFP used a standard measure to target communes throughout the country. In order to make the targeting fair, we used a wide range indicators, which would help to identify problems in different regions. In most cases, WFP was able to identify poor communes using this approach. The drawback however, was a complex index that did not clearly link with causes of food insecurity. One method for targeting used in other countries by WFP is to describe major types of household economies of the people who live there. Areas are then grouped into “Food Security” zones. Grouping districts by types of food economies accomplishes several things: Allows WFP and MRD to target based on appropriate indicators for a particular group. Makes opportunities and problems about food security more clear and understandable to people Helps to simplify targeting procedures, by reducing the number of indicators for each group. Defines a stronger relationship between monitoring events that change food security, and more chronic problems faced in “normal” years. The vulnerability analysis for targeting in 1998, is based on a number of sources: The SESC/MICS Survey: Household survey of 5,600 cases, with nutrition and income information. UNDP/ADB/Ministry of Planning The 1997 Crop Survey: Commune level statistics of rice and other crop production. WFP/FAO/Ministry of Agriculture The MRD/WFP 1997 Commune and Village Rapid Appraisal: 2751 villages looking at a variety of indicators Geographic Information Systems: Includes land use, roads, administration boundaries. Department of Geography/WFP/UNDP/FAO et al. 1)Lowland rain fed areas: The majority of the population rely on a single non-irrigated wet season rice crop as a major food and income resource; small independent land holdings; income is supplemented by a variety of seasonal activities. Terrain is characterised as relatively flat and under extensive cultivation. 45 districts, 2.9 million people 2)Scrub/contract labour: People rely mainly on degraded-forest resources and wage labour for income. Limited cultivation of rice, insufficient to meet annual needs. Landless households are commonly found in these areas. Vulnerable to reduction of forest resources through exploitation, and isolation from markets and major roads. 24 districts, 1.2 million people reside in these areas. 3)Riverine: The majority of people rely on cash crops, floating or dry season rice, and fishing for food security and income. Reside next to major rivers or in communes adjoining the Tonle Sap. Approximately 1.7 million people in 28 districts. 4)Urban/Market: Rely on cash income jobs and small business in urban government centres. 17 districts, estimated at 1.3 million people. 5)Forest: Rely mainly on forest products for food and income. Vulnerable to access to forested areas. Very low densities, less than 8 people per square kilometer. 37 districts, 450,000 people. Mixed: Forest/Rice: Despite meeting the rain fed rice criteria, 4 districts, representing 229,000 people also have over 20 % of the assessed households report cutting logs for income. Forest/Riverine: Over 20% of the population are engaged with forest activities, over 75% of the land area is in forest, and the population meet with the criteria of the Riverine districts. 3 Districts, totalling 122,000 Forest/Scrub: In 9 districts, 230,900 people, families heavily rely on both degraded forests in the vicinity of their villages, and on forest resources within their districts. Rice/Scrub/Forest: 104,500 people in 3 districts meet the criteria for all three classifications. Unclear: In 7 districts, representing 565,000 people, households no definite pattern of food security emerges, or districts straddle several different zones. Criteria for Food Economy assignments Lowland Rain fed meet at least 3 of the four criteria below: =>0.8 hectares of rice land to rice farming households =>250 kilograms of paddy production per farming person =>75% of the households in the district farming rice =>30% of village points inside rice lands Sources: Summary of 1996/1997 crop findings and village-land use analysis. (weighted by number of rice farming households in 1997. See lowland1.sav) Riverine: => 50% of the population reside within 0.5 kilometre of major rivers identified on the RIV_UTM map, or in communes adjoining the Tonle Sap Lake. =>.5 hectares of dry season, floating rice, and other crops per farming household. Districts report at least 30 % of the hectares for rice farming is dry season farming, or secondary cash crop. Sources: GIS: a 500 meter buffer was defined for each river channel, using the major river map, digitized from 1:250,000 maps. Communes were selected next to the Tonle Sap, digitized from 1:50,000 maps. The Communes were combined into a single polygon. Villages from Phum_01 were selected using the polygon tool. Not in buffer: 0; Tonle Sap: 1; Rivers: 2. Populations within buffer were summed, and grouped by district as a field, (Rivbufpop). WFP/FAO/MINAG Crop Survey: Communes were calculated for number of cultivated hectares in dry season rice, floating rice and cash crops. (drycrop), Another field, planpcdry, is calculated as drycrop/plan_all (all cultivated rice land), indicating the percent of the crop in the commune falling in these categories. (Extent of farmed land) Scrub/Contract Labour: =>50% of the people reside in villages in shrub/grasslands or upland crop land classifications AND/OR: >50% of the communes were included in the 1997 targeting exercise, with =>40% of households report selected forest products for income activities in 1996. (Bamboo, Thatch, Charcoal, firewood, grass) (Sources: GIS Simplified Land Cover Map, from 1992/1993 Landsat; Village points with populations were assigned LU class, and calculated as a percentage of total village populations by district Village assessment: districts have at least 50% of the communes surveyed were calculated for the following income activities: Thatch, bamboo, firewood, charcoal, and grass products. Because of known underreporting tendencies on income, we used a 40% threshold.) Forest: =>50% of the land area of the district is primary or secondary forest. AND/OR =>30% of the households in village assessment report log cutting or cutting logs for income in 1996. =<8 persons per square kilometre average density. Urban/Market: Urban Province Centers/Urban Phnom Penh UNFPA designation during the Demographic Survey Mixed: Meet more than one food security zone criteria. Unclear: Meet none of the food security zone criteria Targeting districts and communes Once Food Economy zones were defined and reviewed by field staff, the VAM Unit ran a number of exploratory analyses to determine the best sets of predictors for the different zones. In this step we found that rice crop statistics were a powerful determinant to outcomes in lowland rice areas, but not in other areas. Cash income sources and reliance on high risk coping strategies were closely linked to scrub land zones, while distance from markets and major roads were more closely linked to riverine outcomes. In all zones, being adjacent to a national highway was one of the single most important factors to both nutritional and expenditure outcomes. After the analyses, a basic set of targeting indicators at district level were selected and reviewed by the Programme staff. Findings were processed, and field staff again reviewed and discussed the results. Once districts were ranked and selected, the results of the 1997 Commune Assessment were used to prioritise communes for targeting. In order to ensure that targeting is based on current conditions, a set of criteria have been adopted to update commune information, which would show significant changes in reliance on high risk coping strategies, changes in production, and sudden changes in sources of income.