Cash transfers and vouchers for conflict-affected households in the Democratic Republic of Congo: evidence from UNICEF’s ARCC II programme

In the Democratic Republic of Congo (DRC), evidence has been at the heart of an expansion of cash and voucher assistance in recent years. From the first pilot of cash transfers in 2006 to a randomised comparison of cash and vouchers in 2012, humanitarian research and practice have firmly established that cash and vouchers can be appropriate and effective at meeting humanitarian needs, even amid the severely limited financial infrastructure in eastern DRC. Key studies include S. Bailey and S. Walsh, ‘The Use of Cash in Emergency and Post-emergency Non-food Item Programmes: A Case Study from the Democratic Republic of Congo’, Journal of Humanitarian Assistance, 2007; J. C. Aker, Examining Differences in the Effectiveness and Impacts of Vouchers and Unconditional Cash Transfers, Final Report, Concern Worldwide and Tufts University, 2012. The growth of cash and vouchers arguably constitutes the most significant evolution in humanitarian aid in the DRC in the last decade. However, while vouchers are increasingly common, cash transfers still account for only a very small proportion of assistance. S. Bailey, Humanitarian Cash Transfers in the Democratic Republic of Congo, Working Paper 507, Overseas Development Institute, 2017.

In order to promote learning and evidence on cash transfers and vouchers, the UN Children’s Fund (UNICEF) included a research component in its Alternative Assistance for Communities in Crisis II (ARCC II) programme, funded by the UK Department for International Development (DFID). ARCC II provided $2,781,660 in cash and vouchers to 23,480 emergency-affected households in eastern DRC. Research conducted by the American Institutes for Research (AIR) examined how cash and vouchers influenced household spending; the impact of cash transfers and vouchers on food security, asset ownership, the use of basic services and household resilience; and whether the number of transfers received or the gender of the registered recipient influences households’ decision-making and outcomes. J. Bonilla et al., Humanitarian Cash Transfers in the Democratic Republic of the Congo: Evidence from UNICEF’s ARCC II, AIR, 2017, www.air.org/resource/humanitarian-cash-transfers-democratic-republic-congo-evidence-unicef-s-arcc-ii-programme.

Programme and research design

In 2011, UNICEF undertook pilots providing unconditional cash transfers to emergency-affected households – activities that constituted the beginning of the ARCC programme. While the first iteration of ARCC (known later as ARCC I) mainly provided vouchers for non-food items in ‘fairs’ organised by aid agencies, it also included three pilot projects – one providing vouchers that were redeemable with merchants in existing local markets (rather than fairs) and two distributing unconditional cash transfers.

The subsequent ARCC II programme, also funded by DFID, ran from March 2013 to September 2015. Its primary component was the provision of cash transfers or vouchers to conflict-affected families to enable them to basic needs, access services and pursue livelihoods. Most families (63.5%) received the entire amount in cash transfers, 8% received vouchers and 28.5% received a combination of both. Both the cash transfers and vouchers were ‘multipurpose’, meaning that the programme sought to enable families to use the transfers to obtain the various goods and services that they needed, rather than confining the objective to a single humanitarian sector (e.g. non-food items, shelter, food).

The AIR study took a mixed methods approach employing quantitative and qualitative analysis. The quantitative analysis used data collected by NGO partners to compare ARCC II beneficiaries who had already received their cash or vouchers (Phase I) with ones who had yet to receive aid (Phase II). The Phase II beneficiaries at baseline served as a comparison group to analyse how Phase I beneficiaries’ lives might have changed over time had they not received the cash or voucher assistance. The qualitative analysis used data collected by AIR as well as by UNICEF’s partners during ARCC II. Limitations of the methodology include potential bias in data collection from implementing partners, the lack of a control group, possible unobserved differences between Phase I and Phase II beneficiaries, and differences between Phase I and Phase II quantitative instruments and data collection timing. Unless otherwise stated, the results discussed below are not disaggregated by those receiving cash, vouchers or a combination, and rather refer to the changes experienced across all of the beneficiaries.

Study findings All ARCC II research findings are adapted from Bonilla et al., Humanitarian Cash Transfers in the Democratic Republic of the Congo.

Perhaps the greatest advantage of cash transfers and vouchers (in cases where people have diverse choices for redeeming vouchers) is that households can purchase what they need most. Beneficiaries of ARCC II spread their purchases across a wide range of items and services. Table 1 compares some of the key expenditure categories, the total amount of money (across all families) that was spent on different areas and the total percentage of families who spent some portion of their money or vouchers on this category. The largest purchases were clothing (21% of total expenditures) and household items (20%), followed by livestock (13%), food (9%), education (9%), health (6%), land (4%), housing (3%), debt repayment (2%) and savings (2%). Very little money (0.07%) went to ‘anti-social purchases’ such as alcohol, tobacco or gambling – a finding that is consistent with studies elsewhere. Evans and Popova reviewed 30 studies to explore the impact of cash transfers on the purchase of ‘temptation’ goods. Almost without exception, studies find either no significant impact or a significant negative impact of transfers on temptation goods. D. Evans and A. Popova, Cash Transfers and Temptation Goods: A Review of Global Evidence, Policy Research Working Paper, The World Bank. Families spent the cash and vouchers on essential needs, accessing basic services including education and health, and making investments in livelihood activities – all of which were in line with the objectives of the ARCC II programme. As shown in Figure 1, households receiving cash had more diverse purchasing patterns than those receiving vouchers. Households receiving cash also spent money on goods and services that could not be directly purchased with vouchers, such as debt repayment, land and housing.

Table 1: How families spent cash and vouchers

Figure 1: Distribution of expenditure according to the type of transfer

Source: Bonilla et al., 2017

Food consumption and food security indicators improved as a result of the cash and voucher assistance. Food Consumption Scores improved on average by 30%, and there was an increase in the proportion of households above the ‘poor’ consumption threshold from 59% to 79%. Beneficiary households increased the diversity of foods that they ate – especially meat and dairy. They were also 53 percentage points less likely than the comparison households to have gone to bed hungry, and 17 points less likely to have gone a whole day without food.

As a result of the assistance, ARCC II beneficiaries earned more, saved more, increased ownership of livestock and had more/higher-quality household goods. They earned 47% more (an additional $7.89) than comparison households from their top three income sources. The number of households earning income from their own farms increased by 12 percentage points. More beneficiaries were able to save (16% compared to 5% of comparison households); households with debt decreased by 10 percentage points.

Both quantitative and qualitative data analysis found that households frequently spent a portion of the transfer on their children, with about one in four making purchases for children’s health and education, and 35% purchasing clothing for children. Some outcomes for children improved in relation to the comparison households. ARCC II increased children’s access to healthcare, with fewer beneficiaries (by 21 percentage points) having to forego children’s heath visits due to lack of money. There were uneven results for primary school enrolment, producing an effect for boys but not for girls. Boys experienced a 13 percentage point increase in enrolment – a notable increase given that only 55% of the boys in the comparison group were enrolled. There was, however, no apparent significant impact on enrolment for girls (49% of school-age girls in beneficiary and comparison households were enrolled).

Indicators associated with household resilience (i.e. ability to withstand shocks) improved. An index was created based on ARCC II data and indicators that are commonly associated with resilience, including livestock ownership, children accessing healthcare, access to land, membership in a community group, number of shocks experienced and the household head having attended some secondary school. For both phases’ beneficiaries, the resilience index scores improved from the beginning of the programme.

Effects on community and intra-household relationships were reported as mostly neutral or positive. Beneficiaries saw the transfers as either not changing or improving relationships in their communities because more households had resources – lessening their need to ask for help and increasing their ability to help others. While non-beneficiaries complained about not receiving the assistance in qualitative data collection, there were no reports of these issues escalating into conflict. Several beneficiaries reported feeling an increased sense of respect from community members. Project monitoring found that relationships within beneficiary households did not change (55%) or improved (44%). Qualitative data collection found some anecdotal reports of problems related to husbands being perceived by wives as having made irresponsible purchases.

ARCC II partner MercyCorps randomly selected some households to receive a single transfer of $120 and other households to receive three transfers over three months ($60 followed by two instalments of $30 each). The number of instalments had little impact on expenditure patterns and none on outcomes. Whether the cash transfer was received as a lump sum or in three instalments did not alter income, savings, food security, or children’s access to services. Beneficiaries preferred a single one-time transfer because, with the large, single transfer, they could better manage their money. One instalment also significantly simplifies logistics and reduces delivery costs.

To explore whether the gender of the registered recipient affected spending and outcomes, ARCC II partner Solidarités International randomly varied the registered beneficiary to be a male household member, a female household member or a member chosen by the household. The gender of the registered recipient had little impact on expenditures and none on outcomes. Among the three groups, there were no statistically significant differences in outcomes of interest, such as income, savings, food consumption, asset possession or children’s access to healthcare and education. While some expenditure varied, the results support letting the household choose who receives the transfer.

Conclusion

Research has been crucial in promoting a better understanding of cash transfers and vouchers in DRC – their comparative benefits and costs, how they affect the lives of the people assisted and what those people prefer. The research on ARCC II adds to the evidence base by unequivocally confirming that cash transfers can be an appropriate and effective approach in DRC, and demonstrating how cash offers more flexibility for beneficiaries compared to vouchers. The improvements in household possessions, food consumption, income, children’s access to health treatment and indicators related to resilience demonstrate the potential of cash-based responses to meet basic needs, increase access to existing services and even support the ability of households to manage future shocks. The research in no way suggests that cash transfers or vouchers are all that people affected by crisis need, but it does highlight the opportunities to provide a dignified alternative to in-kind assistance that can meet needs related to various humanitarian sectors. It also underscores the importance of always undertaking analysis on the best type of response in a given area, based on needs, markets, financial infrastructure and risks. By continuing to improve evidence, practice and partnerships through programmes such as ARCC II, it is hoped that more and more people in DRC will receive cash transfers and vouchers when they are the best response option.