Action Against Hunger (ACF USA) has been working in Uganda since 1980, with operations in Karamoja and Northern Uganda. ACFs food security and livelihood operation in Otuke District (former Lira District) has changed since 2005, in line with the changing context from displacement to return to reconstruction over the last five years. In particular, as needs have diversified and markets have been re-established, ACF has increasingly turned to market-based and cash-based interventions, as shown in Table 1.
This article focuses on the LEARN-1 project implemented by ACF in Otuke District from February 2009 to January 2010. The project was designed to support the return and livelihood recovery process in Northern Uganda.
For ACF, the LEARN project was an opportunity to apply and expand its growing technical expertise in cash-based interventions, extending the cash transfer approach to Uganda and more specifically to Otuke District, where voucher-based projects had been under way for several years. In total, 34 villages in nine parishes were selected. Fifteen hundred of the most vulnerable households were given two cash transfers each totalling 570,000 Ugandan shillings (USh) ($285), spaced several months apart. There were four targeting criteria: human (disability, elderly, chronically ill, family members in feeding programmes); economic (loss of productive assets, loss of labour opportunities, debt, no access to savings/credit); socio-political (no family support, no external assistance, returnee/internally displaced); dependency ratio (number of family members below 14, above 60 years of age or chronically ill per economic productive family member); and suitability for the programme (including not moving away from the geographical area, and willingness to complete preparatory training).
A Memorandum of Understanding (MoU) was signed with the Equity Bank branch in Lira town to facilitate the cash transfers. Equity Bank opened bank accounts for beneficiaries free of charge and conducted training sessions in the field on the use of the accounts. Although there were plans for Equity Bank to install several points of sale devices in Otuke District to facilitate access to the accounts, this was not in the end possible. Instead, the bank brought cash to the field using security escorts.
Following initial training, including the development of a household action plan for using the cash, beneficiaries received an initial transfer of USh250,000 in late July or early August 2009. Between the first and second disbursements, training was conducted in the four main activity areas selected by beneficiaries (crop production, livestock, animal traction and small business management). The second cash transfer was made between late November and early January. The overall cash grant was split into two transfers to facilitate monitoring activities between the transfers, and to ensure that funds from the first transfer were not misused for illegal or anti-social purposes, which would have led to exclusion from the programme and no second transfer.
Besides the usual monitoring of baseline, post-distribution monitoring and a final evaluation in June 2010, a graduate student from the Institute des Regions Chaudes of Montpellier SupAgro in France spent three months with ACF researching the livelihoods of programme beneficiaries in Summer 2009. A consultant from the Cash Learning Partnership (CaLP) also visited Lira to assess the impact of the cash transfers on local markets in November 2009.
Impact on food security and livelihoods
The impact of unconditional cash transfers is much more difficult to predict than that of in-kind or voucher-based food security and livelihood projects. One of the main advantages of unconditional cash transfers is that beneficiaries have complete freedom to spend aid money on what is best for them, rather than having to accept what the implementing organisation or donor thinks is best for them. From past programme experience and beneficiary feedback, ACF thought that households would invest primarily in seed and other agricultural inputs, and in small businesses. Indeed, had this been an in-kind distribution project ACF would probably have distributed seed and income generation activity (IGA) kits. Instead, the majority of beneficiaries chose to buy livestock. Approximately USh650,000,000 ($325,000) was spent on livestock purchases. In total, the average household spent USh480,000 ($240, or 84% of the total grant) on productive assets, and UShUGX90,000 ($45, equivalent to 16% of the total grant) on immediate needs. Over half of the money used for immediate needs was spent on food (54%), with smaller percentages going on medical bills (11%), school fees (9%), household items (7%) and debt repayment (4%).
The impact of the project on the livelihood assets of households was generally significant, as most invested in livestock. This accelerated a long-term household recapitalisation process, which is explained in Figure 1.
The project also had a significant impact on local non-beneficiaries particularly medium-sized farmers and small traders from whom beneficiaries purchased many of their items. The sustainability of the project was good, largely because beneficiaries chose to invest in long-term productive assets. Roughly 70% of households used some of the cash to purchase food, but quantities were quite small (USh50,000 ($25)). Most of the food purchases were made with the first transfer, which came during the annual hunger gap, and hence the food purchased had a transitory effect on households food security.
Impact on markets
Aside from the impact on beneficiaries themselves, one of the frequently cited advantages of cash-based interventions is the potential positive impact they can have on the local economy. The Cash Learning Partnership (CaLP) conducted a study on the impact of the LEARN-1 cash transfers on local markets in Otuke and Lira District following the first cash disbursement. The study concluded that at least 50% of the money from the first disbursement passed directly to medium-sized farmers in Otuke, who were the main source of goats for beneficiaries. Aside from goats, many of the other items purchased by beneficiaries were bought outside of Otuke. Although some cattle were purchased in Orum market from traders bringing cattle from Karamoja, most were obtained in livestock markets elsewhere in Lira District (Apala, Amach) or in the neighbouring Teso sub-region. Ox ploughs and income-generating items were mostly purchased from Lira town. Figure 2, which is reproduced from the CaLP study, shows the cash flow following the disbursement. Whereas in-kind and voucher projects result in cash going directly to large traders, with direct cash transfers much of the cash passes first through medium farmers and small traders. This intermediate step represents a significant secondary impact of the project on local non-beneficiaries.
Due to the open nature of the Ugandan economy and the relatively small influx of cash in relation to the economy as a whole, there was no inflationary impact on medium- and long-term local prices. However, the CaLP found that there was a short-term rise in prices, which it refers to as flash inflation. This occurred in the local weekly markets in the trading centres of Otuke during the 12 weeks following the disbursement.
Unconditionality and grant size
One of the main advantages of unconditional cash transfers is that beneficiaries, not implementing organisations, decide how to spend the money. The LEARN project clearly demonstrates that cash transfers enable recipients to quickly adapt to changing circumstances and needs.
Deciding on the size of the cash transfer is one of the most important decisions in the design of a project. The amount of cash provided needs to be large enough to have a significant impact on household food security and livelihoods. In this case, the figure of USh570,000 for the combined transfer was arrived at based on local market prices and estimates of household needs. Greater than the entire annual income of some households, the transfers enabled purchases of productive assets (e.g. starting an income-generating activity and buying livestock), while also covering immediate expenditures.
Nevertheless, cash transfers should not be seen as the solution to all of a households problems, but rather as an effective way to help them restart key economic activities from which they can make profits and grow and expand in the future. The minimum amount of cash required for households in Otuke to become food secure, as indicated in Figure 3 as the Critical Threshold. This is estimated at USh160,000 ($80), which is enough to buy four goats. To enable households to launch a significant productive activity, USh250,000 ($125) would be needed. Given current prices, an additional USh100,000 ($50) would be required to help cover immediate needs, such as food and medical bills. Taking into account potential variations in beneficiaries situations and market prices (with an additional margin of USh50,000), USh400,000 ($200) would be the amount of cash needed to have a long-term impact on household livelihoods and food security. Comparatively, the transfers were potentially larger than necessary to achieve the above-mentioned objectives of passing the Critical Threshold and attaining food security, hence catapulting the most vulnerable households far ahead in the livestock capitalisation process and turning them in very short order into the big businessmen and employers of their villages.
The evidence from the project shows that the impact of unconditional cash transfers can be significantly greater than comparable in-kind or voucher projects because beneficiaries are free to choose how to spend the cash at the time it is received. Hence the unconditional nature of the grant is a significant additional benefit in itself, allowing recipients the flexibility to respond to changes in their circumstances. Evidence from the LEARN project suggests that even highly vulnerable households will use cash transfers to invest in long-term productive assets, while the flexibility of cash facilitates work with the most vulnerable because part of the transfer can be used to cover immediate needs. When most items purchased are produced locally, cash transfers can bring substantial efficiency gains, even when beneficiaries live far from major markets.
In-kind assistance (or part of it) is often exchanged at below market value for other items the household needs more, lessening its value and impact. Especially in cases where distributions arrive out of sync with the seasonal calendar, beneficiaries often have no choice but to barter the goods received. The impact of unconditional cash transfers can be significantly larger than comparable in-kind or voucher projects because beneficiaries can prioritise expenditure according to their needs and recent or anticipated changes in the local context.
The design of any programme is best supported by a participatory baseline study, with communities potentially participating, which anticipates needs on the ground and helps define the details of the planned transfer, at the time of programme inception. A cash transfer programme needs to define cost calculations and transfer amounts, linked to the programme objectives. The flexibility inherent in a cash transfer allows for situations where fast-changing contexts and beneficiary priorities might diverge from the initial baseline study. Cash transfer amounts should be appropriate to the particular context (e.g. emergency, recovery, development). More work is needed to understand how cash transfers are used in different contexts, and to determine appropriate cash transfer amounts. Cash transfers are not likely to cause general price inflation in an open economy with interlinked markets, though short-term inflation in local markets can have a significant impact on the prices of goods immediately after cash disbursement.
Once the LEARN-1 Norwegian Embassy Consortium evaluation has been completed, more analysis needs to be focused on the direct impact on household-level food security and livelihoods, for instance in terms of income, dietary diversity and nutritional status, and other livelihood indicators. Additionally, given available resources, a comparison should be made between ACFs past in-kind or voucher programmes (seeds and IGAs) and the current unconditional cash transfer programme, and the impact and sustainability of each approach on household food security and livelihoods.
Silke Pietzsch is Senior Food Security and Livelihoods Advisor for ACF-USA, based in New York. Silke would like to thank Erik Engel (ACF Lira FSL Programme Manager) and Hanibal Abiy Worku (ACF Uganda FSL Coordinator) for implementing the programme discussed here, and Obie Porteous, who conducted the LEARN-1 evaluation.
 The main reason for using bank accounts was so that beneficiaries would not need to store large amounts of cash in their homes, and to encourage saving on a more formal level. Local Savings and Credit Co-operatives (SACCOs) did not provide the necessary financial capacity and transparency.
 The proposed amount was actually $250, covering agriculture inputs/IGA equipment, food and other needs. The amount was later increased during the course of the programme due to increases in the budget as a result of the favourable exchange rate with the Norwegian Kroner.