‘Zap’ it to me: short-term impacts of a mobile cash transfer programme

Summary of working paper¹

Hadijatou with her cash, Toro village

In 2010, Concern Worldwide developed a humanitarian programme in response to the 2009/10 Niger drought and food crisis. In an attempt to prevent asset depletion and reduce malnutrition among drought-affected households, the programme provided unconditional cash transfers to approximately 10,000 households during the hungry season, the five-month period before the harvest and typically the time of increased malnutrition. Programme recipients were to receive an average of 22,000 CFA (US$ 45) per month for five months, to a total of US$ 215. In an effort to facilitate the disbursement of cash in remote areas, Concern decided to implement a pilot study across 116 villages in six communes of the Tahoua region.

Three interventions were chosen for the pilot programme. The reference was the standard manual cash intervention, whereby beneficiary households received unconditional cash transfers of 22,000 CFA ($US45) per month. The total value of the transfer over the five-month period was approximately two-thirds of the total annual gross domestic product (GDP) per capita. Payments were made on a monthly basis, whereby cash was counted into envelopes and transported via armoured vehicles to individual recipients. Rather than distributing the cash in each village, a central village location was chosen for groups of 4-5 villages. Programme recipients had to travel to their designated location on a given day to receive the cash transfer.

The two additional interventions were variants of the basic intervention. One of these aimed to reduce the costs of distributing cash to remote, sparsely-populated and in some cases, insecure rural areas. Programme recipients in the second group (zap) received their cash transfer via mobile phone (m-transfer). After receiving the electronic transfer, recipients had to take the mobile phone to a m-transfer agent located in their village, a nearby village or a nearby market to obtain their physical cash. Since less than 30 percent of households in the region owned mobile phones prior to the programme, Concern also provided programme recipients with mobile phones, as well as the zap account, and paid for the transfer charges. The second intervention thereby differed from the manual cash intervention with respect to the transfer delivery mechanism, as well as the provision of the handset and the m-transfer technology.

In an effort to disentangle the impact of the change in delivery mechanism from that of receiving a mobile phone, the third group (placebo) mirrored the manual cash intervention, but also provided a mobile phone. Like the manual cash group, programme recipients received $US45 in physical cash on a monthly basis and had to travel to a meeting point to receive their cash. However, like the zap group, programme recipients also received a mobile phone (which was ‘zap’ enabled), yet did not receive their transfer via the mobile phone.

Prior to the introduction of the programme, Concern Worldwide identified 116 ‘food deficit’ villages in the Tahoua region, i.e. those classified by the Government of Niger as having produced less than 50 percent of their consumption needs during the 2009 harvest. Of these, some villages were prioritised for the zap intervention based upon their population size and proximity to skirmishes near the Niger-Mali border, thereby reducing the sample size to 96 villages. The remaining eligible villages were randomly assigned between the basic (manual cash), placebo and zap interventions. In all, 32 villages were assigned to the cash group, 32 to the placebo group and 32 to the zap group.

Within each food deficit village, household- level eligibility was determined by two primary criteria: the level of household poverty (determined during a village-level vulnerability exercise) and whether the household had at least one child under five years. The number of recipient households per village ranged from 20 to 75 percent of the village population. In all villages, the cash transfer was provided to the woman.

A comprehensive household survey of more than 1,200 programme recipients was conducted in all 96 villages. The baseline survey was conducted in April 2010, with a follow-up survey in December 2010. The research team located over 98 percent of households for the follow-up survey. The household survey collected detailed information on household food security, demographics, asset ownership, agricultural production and sales, mobile phone ownership and usage, uses of the cash transfer and village and householdlevel shocks. A second dataset was collated on weekly agricultural price information from over forty-five markets for a variety of goods between May 2010 and January 2011, as well as the date of each cash transfer in each village. These data were used to test for different effects of the cash transfer delivery mechanism (zap or manual cash) on local market prices.

The average per recipient cost over the life of the project was US$12.76 in cash/placebo villages and US$13.65 in zap villages, or US$0.90 more per recipient. While there was a range of benefits from the zap intervention, the research focused on two in particular for the cost-benefit analysis: the monetary value of the reduced opportunity costs of programme recipients’ time (a value of US$0 .91) and the increased cultivation of cash crops. Using average household okra production and the market price for okra during the programme period, the average value of this okra production in zap households would have been US$5. This suggests that the cost-benefit ratio is greater than one, meaning that the additional costs of the zap intervention yielded an equivalent or higher monetary benefit for zap programme recipients. If the programme yields benefits in the longer-term, perhaps by allowing households to send and receive more informal transfers or access formal financial services, this could potentially yield a higher rate of return.

Women in Abala Sani village at a mobile phone training session

An intervention that provided a cash transfer via the mobile phone strongly reduced the costs of programme recipients in obtaining the cash transfer, and reduced the implementing agency’s variable costs associated with distributing cash. This suggests that mobile telephony could be a simple and low-cost way to deliver cash transfers. In addition, those in the m-transfer group bought more types of food and non-food items, increased their diet diversity, depleted their non-durable assets at a slower rate and produced a more diverse basket of agricultural goods. These differences are primarily due to the m-transfer intervention, and not to the presence of the mobile phone, suggesting that a programme that simply distributes mobile phones might not yield the same impacts. These effects appear to be due to the reduced costs of the programme and the greater privacy of the mtransfer mechanism, which are potentially linked with changes in intra-household decisionmaking.

The m-transfer approach may be limited in its application to all contexts. First, it will only be effective in cases where telecommunications infrastructure currently exists, which could limit its utility in remote areas. Second, in areas with high rates of illiteracy - as is the case in Niger - programme recipients might not able use the m-transfer technology on their own, implying that they might need help from other family members, friends or m-transfer agents. This could potentially limit the use of the technology by programme recipients for informal private transfers or in accessing other mobile financial services, but could be beneficial for the household as a whole. And finally, the short-term impacts of the programme might not persist in the longer-term. Despite these caveats, the widespread growth of mobile phone coverage, cheaper mobile phone handsets and m-money services in developing countries suggests that these constraints could be easily overcome. In addition, the benefits of the programme in a context such as Niger - a country with limited investment in power, roads and landlines, low literacy rates and one of the highest rates of financial exclusion in sub- Saharan Africa - suggests that the approach could thrive in less marginalised contexts.

¹Aker. J et al (2011). Zap it to me: The short term impacts of a mobile cash transfer programme. Centre for Global Development. Working Paper 268. September 2011.