Vouchers and fairs as emergency response in DRC

Summary of evaluation¹

Masisi Centre fair

In late 2008, escalated fighting among rebels and the Congolese Armed Forces (FARCD) provoked renewed and widespread displacement in North Kivu province, Democratic Republic of Congo (DRC). In response to the unfolding crisis, Concern Worldwide implemented an Emergency Assistance Project to newly displaced and host families in Masisi territory funded by the Disasters Emergency Committee (DEC) and Irish Aid.

Concern's nine-month emergency project sought to contribute to the livelihood security of displaced and host families in Masisi territory by increasing household access to food, non-food items (NFIs) and water. Concern aimed to identify and assess the needs of 10,000 vulnerable and newly displaced households,² meet emergency NFI needs, improve water access for households in Rubaya, and provide livelihood support for four months for the poorest beneficiaries. A primary activity of the project was creating 'fairs' where beneficiaries could use vouchers to purchase non-food items (e.g. kitchen utensils, pagnes³), seeds and tools from vendors, as well as to pay school fees. Concern opted to use fairs and vouchers rather than distributing NFI kits, to provide beneficiaries flexibility and choice.

An independent evaluation was conducted in August 2009 to assess the extent to which the project achieved its objectives, with particular emphasis on the appropriateness of the voucher activities.

Key findings

The evaluation found that fairs were an appropriate and effective response to the needs created by displacement and return. Assessments accurately identified priorities among households in the intervention area, which laid most emphasis on the replacement of basic household goods lost when they fled. The basic conditions were in place for implementing fairs - traders could supply critical items, authorities supported the intervention and beneficiaries felt that the fairs would be an appropriate way to meet needs and would not put them at risk.

Concern organised fairs with the following basic parameters:

• Beneficiaries had $35 worth of vouchers (in Francs Congolais) and also received two blankets and soap from Concern.
• Traders were pre-selected from intervention areas (as opposed to Goma city) in order to contribute to the local economy.
• Articles to sell (kitchen articles, pagnes, haricot/pea seeds, machetes, hoes, headscarves) were pre-selected based on focus group discussions (FGDs) with women.
• Prices of articles were fixed based on market research and negotiations with traders.
• Primary school fees could be paid to headmasters.

The fairs provided beneficiaries with access to key goods and services in a way that enabled them to make choices according to their own priorities. Most beneficiaries were highly satisfied with the process. The majority preferred the fairs to receiving pre-packaged NFI kits, cash or vouchers to use in the local markets because the fairs enabled them to pay school fees, provided them with choice and were safer than receiving cash or vouchers to use in the local market.

The option of paying school fees was an especially popular aspect of the intervention. One-fifth of beneficiaries used vouchers to pay school fees and 27% of those beneficiaries paid for more than one child or trimester. For parents with children in primary school, school fees were priority expenditure that they were having trouble meeting because fighting and displacement had interrupted their livelihoods. The payment of school fees enabled children to continue their studies uninterrupted, as they were often sent home from school when parents failed to pay. The inclusion of school fees tipped many recipients towards preferring fairs compared with receiving in-kind assistance.

Addressing certain issues would have further increased the effectiveness of the fairs. In spite of substantial sensitisation efforts using community members, a small number of beneficiaries simply did not grasp the process. In rare cases, they exchanged all of their vouchers for only one or two items. The prices of certain goods in the fairs were higher than market prices. The selection of articles should have been wider and should have included used clothing, which was in high demand by beneficiaries. There were some cases of articles running out at certain times. The lack of small voucher denominations reduced flexibility because beneficiaries could not pay for articles with precise amounts and traders did not provide change in the form of cash or vouchers. Beneficiaries therefore had to purchase multiple items with one trader or receive change in the form of soap or salt.

In terms of efficiency and cost effectiveness, using vouchers required more planning and was slightly more expensive compared with in-kind distributions, but had the benefit of providing more choice to recipients and supporting local traders. Sensitisation with beneficiaries and traders, combined with the creation of fairs that served around 700 households per day, took more time than the well-established approach of procuring and distributing NFI kits (up to 1,000 households per distribution day). The fair approach was marginally more expensive since use of local traders meant it was not possible to benefit from savings inherent in bulk purchasing through competitive procurement processes. However, working with local traders directed profits to businesses in the intervention areas as opposed to those in Goma or outside of DRC. There is also widespread recognition among nongovernmental organisations (NGOs) that beneficiaries often sell certain NFI kit items to meet other priorities; the use of vouchers probably discouraged this resale by providing choice and the option of paying school fees.

In targeting the assistance, Concern used a participatory exercise where a committee of community members ranked displaced and returnee households (some of which were hosting displaced households) into categories of 'poor', 'very poor' or 'poorest of the poor'. Concern then targeted assistance to the latter two categories. The result was high levels of understanding, satisfaction and inclusion in Rubaya and Bihambwe, where upwards of 95% of households were targeted and limited inclusion/ exclusion errors were reported. Given that the assistance provided did not vary between the bottom two wealth categories, the targeting process could have been simplified by establishing one set of criteria for inclusion rather than ranking households into three wealth groups. In Matanda, where the 'poorest of the poor' were targeted because remaining resources could serve less than half of the population, the process was complicated by a lack of transparent information on the targeting criteria and the inherent difficulties of targeting assistance to only a portion of the population amid similar levels of need.

Bwira Sancha and her daughter from Luibo, Masisi Territory, with a cash voucher

Because of a one-month project suspension following an attack on a Concern base, Concern decided not to implement the livelihoods support component of the project through a monthly cash or voucher transfer. For the water activities, FGDs and key informant interviews confirmed that water access was a problem in Rubaya, as taps were open only for two hours each day. Following Concern's linking of a spring with an existing tank, water access increased significantly. Leaders and focus groups reported that some taps were breaking because of the stronger water pressure in the tank.

Main conclusions and recommendations

The overarching objective of the project was to contribute to livelihood security of displaced and host household families through increased access to food, NFIs and water. While it is difficult to determine impacts on livelihood security, access to NFIs, agricultural inputs, school fees and water contributed to livelihoods by meeting important basic needs, supporting agricultural livelihoods for those with access to land and reducing household expenditure. There were certainly very positive livelihood impacts for the 17 traders who participated in fairs, who made profits that far exceeded their normal activities. Increasing the value of the voucher would have further augmented support to livelihoods, but at the evident cost of reaching fewer beneficiaries.

Providing choice to recipients was a clear advantage of the fairs compared with more traditional forms of emergency assistance where assistance commodities are distributed. Bringing in more traders would have made choice wider and more consistent. Humanitarian agencies and donors should take note of Concern's inclusion of school fees in this project. In this setting, it was a safe and appropriate way to support education and livelihoods by reducing household expenditure.

High prices of articles in the fairs were far and away the largest complaint of beneficiaries. Prices were higher than 'normal' market prices because of a combination of three factors: certain items were of a different quality to those in the local market, traders did not always provide change to beneficiaries at the fairs, and despite Concern's efforts to set prices at or below market level, prices were not accurate for all items. Taking measures to ensure lower prices in fairs would increase both efficiency and effectiveness.

The emphasis on participation throughout the project had a number of positive benefits. The thorough sensitisation process using community members resulted in a general high level of understanding among beneficiaries. To respond to the rare but important exceptions of beneficiaries who did not understand the process, Concern could have instructed sensitisation committee members and local leaders to identify people who might have trouble understanding the process and encourage people to bring a 'helper'.

Staff showed noteworthy skill and flexibility in quickly adopting a voucher approach, which was new to everyone involved. Concern's successful implementation of fairs shows that organisations do not need 'cash and voucher experts', but rather talented staff who can use their emergency assessment and project management skills to incorporate cash-based responses in their interventions. Having training sessions and centralised resources on cash-based responses readily available to key staff would facilitate this process, as would exchanges with organisations conducting similar interventions. The lessons from this intervention should serve as source of learning within and outside of Concern, including with agencies in the DRC engaged in similar voucher responses.

¹Baily. S (2009). An independent evaluation of Concern Worldwide's emergency response in North Kivu, Democratic Republic of Congo. Responding to displacement with vouchers and fairs. Humanitarian Policy Group, Overseas Development Institute, London September 2009. Download report at: http://www.odi.org.uk/resources/download/4195.pdf

²The number of beneficiaries was subsequently reduced from 10,000 to an estimated 8,000 as other non-DEC funding initially allocated could not be carried forward beyond 31 December 2008.

³A piece of material used by women as clothing or tailored to make clothing.