Community-based targeting: responding to humanitarian needs in the Democratic Republic of Congo

During humanitarian crises, aid workers must find efficient ways to help the vulnerable meet immediate needs. Because resources are finite, targeting those most in need is often a crucial part of this exercise. However, commonly used targeting strategies in livelihoods interventions may not always identify the most vulnerable. Often, vulnerability criteria are predetermined according to aid agencies’ programme objectives rather than recipients’ needs. Whether based on a particular group of individuals or indices, targeting can sometimes fail to identify existing support structures or recognise less conspicuous vulnerabilities.

In this article we argue that communities, through a facilitated process, are able to create more complex, nuanced vulnerability criteria. We recognise that there are risks in community-based targeting approaches, including local elite capture and increased tensions within communities. Community-based targeting can also be time-consuming, and can take time away from other productive activities. However, it is possible to mitigate some of these risks through facilitation and verification mechanisms.

This article discusses a community-based targeting approach piloted by the International Rescue Committee (IRC) in the eastern Democratic Republic of Congo (DRC). This activity was part of the DFID-funded Tuungane programme, a community-driven reconstruction programme implemented by the International Rescue Committee (IRC) in partnership with CARE International. The programme was suspended in April 2013 following the violent occupation of North Kivu province by the M23 armed group.

In response to the deteriorating food security and livelihoods situation, particularly in Rutshuru Territory, the IRC launched a Livelihoods Emergency Response Initiative, a cash transfer programme that delivered around $1.4 million to 10,000 households between August 2013 and April 2014. As part of the initiative the IRC maintained key principles of its larger Tuungane programme, notably the promotion of good governance principles and community participation, by developing the Participatory Process for Categorization and Allocation (PPCA), a community-based approach to targeting the most vulnerable households facilitated by Village Development Committees (VDCs).

Community-based targeting: the Participatory Process for Categorization and Allocation

In the PPCA approach, the IRC guided VDCs through an eight-step process, carried out over six days, in which they identified and categorised all households in the community by level of general need. Each VDC decided what amount to allocate per household, with the highest allocations going to the most vulnerable, and progressively smaller amounts to households in less vulnerable categories.

The eight steps were as follows:

1. Orientation: IRC gave an overview of the PPCA process to Village Development Committee (VDCs).
2. Creation of a household list: Using a method of their choosing, the VDCs registered each household in their community. This list of households was posted publicly for at least 24 hours to allow for public verification.
3. Validation of the household list and creation of criteria: In the first of two general community assemblies, the list of households was discussed and adjusted if necessary. The final list was validated by general public vote and signed by the members of the VDC, local chiefs and an IRC staff member. In focus groups (separated into men, women and youth), community members developed criteria to categorise households into one of four levels of general need, from most vulnerable (‘Category A’) to least (‘Category D’) (see Box 1). When the groups reconvened in the general assembly, these criteria were discussed, harmonised and validated in a general vote.
4. Categorisation of households: The VDCs, often with the support of additional members of the community, assigned each household to a category according to their level of need using the criteria generated by the assembly.

1. Verification by the IRC: IRC staff visited a sample of households across categories to confirm that the categorisation did not favour elites.
2. The household list by category of need was posted for further public scrutiny.
3. In the second general assembly, the final list was validated through a general public vote.
4. Allocation of funds by category: During the second general assembly, IRC staff met privately with the VDC to give them the total dollar amount allocated to the community, based on a combination of baseline Food Consumption Scores (FCS) and estimated population. The VDC then decided what specific amount to allocate per household per category. For example, the households in the neediest category (Category A) might be allocated $160 each, the households in the next needy category $130 each, and so on. The VDC then returned to the assembly to discuss their allocation plan and put it to a general public vote.

While this process demanded a considerable amount of time from the community, once staff members were familiar with each step they were able to guide the VDCs through it with little difficulty. A team of 10–12 staff took 32 communities (almost 10,000 households) through the process in less than a month.

Assessment findings

The initial feedback from participating communities was so positive that the Tuungane team requested an assessment to determine whether the PPCA was helpful in identifying the most vulnerable and promoting good governance principles of participation, inclusion, accountability and transparency.

The assessment found that the PPCA process had successfully targeted the most vulnerable. Survey responses and focus group discussions during the assessment provided a wide consensus that the most vulnerable households were included in Category A (the most vulnerable), while the better-off households were excluded. It also found that communities created multi-faceted vulnerability criteria, rather than the simple individual characteristics normally used by aid agencies. For example, instead of classifying all widows as ‘vulnerable’, communities distinguished between widows with and without assistance. These composite criteria would be difficult to create without a thorough understanding of the individuals concerned and the dynamics of a particular community. Locally-determined categories also seem to take into account earning potential (e.g. the ability to plant) and other sources of support, such as help from adult offspring, that standardised, agency-led tools are generally unable to capture.

Third, the process was successful because it was community-led, with a high level of community participation and inclusion in the PPCA process. During the assessment survey, a random sample of respondents was asked if they received assistance in the livelihoods intervention. The purpose was to check whether the survey team would encounter some community members who should have been on the beneficiary list, but were left out. Survey results indicate that 95.8% of respondents received items in fairs and/or distributions, while the 4% who did not were returnees who came back after the beneficiary list had been created. Survey results indicate that 87% of respondents (336 out of the 386 respondents who received items in the fairs and deliveries) participated in at least one general assembly.

Because it was a community-led process, there was widespread community knowledge of the criteria. Of the community members surveyed who received items in the fairs and/or deliveries, 92% (354 out of 386) could name at least one of the criteria used to categorise households in the neediest category. A majority of focus group respondents said that they preferred differential voucher amounts depending on household needs, rather than all households receiving the same voucher amounts or items.

According to community members who participated in focus groups, VDC members were well-placed to categorise each household by level of vulnerability because VDC members are themselves part of the community. As such, they know the other community members and their needs. According to VDC members, the categorisation process helped them gain leadership capacity and better knowledge of community members’ needs.

Fourth, the PPCA approach required a significant time investment, particularly from VDC members. Compared to similar distributions, the IRC estimates that the PPCA approach demanded a time commitment from village committee members that was ten times greater than for more traditional, agency-led initiatives (the time commitment required from NGO staff was about the same). Even so, VDC members in general said they felt that this was time well spent because it was an opportunity to steward resources on behalf of their communities.

Fifth, the approach also requires flexibility. When facilitating truly community-led approaches, agencies cannot predetermine define vulnerability, which may differ from expectations and donors’ specific priorities. Only because the IRC and its donor, DFID, maintained flexibility and placed decision-making power in the hands of communities was the project able to use a truly community-led, participatory approach.

A promising approach

The assessment findings show promise for the capacity of communities and community leaders to lead the targeting of vulnerable households. Communities created multi-dimensional vulnerability criteria and the categorisation process was generally perceived as fair, equitable and accurate. The fact that the beneficiary list and the criteria for categorising households were developed by the community meant that community members were better informed about the process and had more buy-in in the process. In addition, the PPCA approach helped Tuungane and the VDCs to maintain a continuity of support to communities despite prevailing insecurity.

Merry Fitzpatrick is a Researcher with the Feinstein International Center at Tufts University, Marie-France Guimond is a Research Specialist at the International Rescue Committee (IRC) and Guillaume Labrecque is a Governance Advisor at IRC.