Save the Children’s Emergency Cash Transfer Programme in Myanmar
by Sue Mark, Save the Children in Myanmar May 2009

In the early hours of 3 May 2008, Cyclone Nargis made landfall in the Ayeyarwaddy Delta in Myanmar. Official figures released by the Post-Nargis Joint Assessment (PONJA) in July 2008 put the death-toll at 84,537, with 53,836 missing and some 2.4 million severely affected. Damage from the cyclone has been put at over $4 billion. Prior to Nargis, there was limited experience in using cash as a form of intervention for relief or development in Myanmar. Nargis presented an opportunity to use cash transfers, given the immense shock to livelihoods, resilient markets and the logistical difficulties in getting goods to the remote villages of the Delta. Although the Myanmar government generally does not view cash transfers favourably, agencies have been allowed to distribute cash for very clearly defined objectives. This article describes Save the Children in Myanmar (SCiM)’s cash transfer programme.

Outline of the programme

In line with the priority needs identified by affected communities, the programme aimed to help cyclone-affected households to ‘begin to fulfill their right to livelihoods and to reduce dependence on food assistance through the rapid replacement of essential livelihoods assets’.With £667,000 in funding from the Disasters Emergency Committee (DEC), the programme provided 28,000 households with cash grants of 50,000 kyats. Livelihoods Committees were formed in each village to facilitate the delivery of livelihoods projects. Assets supplied included boats, nets, livestock, items to start small businesses and fertilizer.

The programme used a mixture of cash grants direct to households and facilitated procurement of livelihood assets. The choice between these two methods was based on several factors, including whether the local authorities would accept the use of cash as a means of recovering livelihoods, the applicability of cash to livelihoods activities, market access and the potential for cash to cause conflict between targeted and non-targeted communities. Safety and logistics were also considered.

In light of the tensions caused within some communities as a result of household-level targeting in food aid programming, it was decided that the project would give equal levels of assistance to each household in the targeted villages. Villages were selected according to the following criteria.

1. Food security: so that households would not use the assistance to buy food, it was essential that people enjoyed minimal levels of food security.

2. Degree of mortality and damage.

3. Degree of household poverty and vulnerability.

4. Multi-sectoral approach: other Save the Children sectors should be working in the village, to ensure maximum outreach capacity and pre-existing relationships.

5. Overlap: avoid areas where other agencies were doing similar work.

The project was launched in the last week of July 2008, and was completed by the end of September, a little over two months later. The target of 28,000 households was reached across seven area offices in the eastern and western regions of the Delta. This is notable given that SCiM did not have a livelihoods programme or staff in place prior to Nargis.

Evaluation

An internal evaluation of the project was conducted, involving focus group discussions with 17 Livelihoods Committees and 102 interviews with households from poor and very poor wealth groups, including 54 in an area which only used cash grants and 48 in an area which used a mix of cash grants and facilitated procurement. As most households used the majority of their cash grant to buy livelihoods assets, there was no obvious difference between the two methods in terms of outcome for the households concerned. A market survey was also carried out, including interviews with seven traders.

Impact on livelihood recovery

Most people said that the project had had an impact on their livelihood recovery, through investment in essential livelihoods assets. Of the 102 households surveyed, 37 said that they had recovered their livelihoods to a quarter or below of their pre-cyclone levels; 52 reported that livelihoods had recovered to 50% or above their pre-cyclone levels; nine stated that their income levels were 100% or above pre-cyclone levels and four did not respond.

Usage of cash

Livelihoods Committees (LCs) from each of the 17 villages visited reported that nearly all the grants had been used for the purchase of assets to support livelihoods, predominantly casual labour, petty trade and fishing. The household surveys also showed that the majority – 66 households – used all of their grants for the purchase of livelihoods assets. Thirty used most of the grant for livelihoods assets, with the balance going on food, education and health expenses; only six households reported not using their grants for livelihoods asset recovery, instead opting to spend all of the money on food, education and health.

Relationship between food and livelihoods

Given the strong relationship between food security and livelihoods, the survey also looked at the impact of the cash transfer on the quality and quantity of food available. Regarding sources of food, 55 out of the 102 households surveyed reported that their primary source of food was food assistance, followed by fishing, purchase and lastly exchange from farm labour. The cash injection resulted in a moderate increase in the quantity and quality of food: of the 102 households, 58 said that food quantity had remained the same, and 34 said that it had increased. The remaining ten households had no answer or said that food quantity was lower. Regarding quality, 49 households reported that it had stayed the same, and 23 said that it had improved. Nineteen households said that the quality of their rice had deteriorated, and 11 gave no answer. An additional question was added to the original survey to assess whether households receiving food assistance were aware that it would end in the coming few months, and how they would cope. No one asked was aware of this possibility, but neither did anyone ask for food aid to continue, provided that people were assisted with the recovery of their livelihoods either through asset replacement or access to capital in the form of a grant or loan.

Expenditure priorities

When asked about their main expenditures, 90 households said that their greatest expenditure was on food, followed by livelihoods assets, health and education. This data is interesting because it tells us that households were still seeking to diversify their diet or add to the food aid they were receiving by buying food. This finding supports an SCiM food assessment conducted in November 2008, which showed that cyclone-affected households, for the most part, maintained a diet similar to their pre-cyclone one, with some reduction in protein, meat and eggs, and that food aid only made up between 30% and 70% of total household food consumption, as targeted households shared their rations with others not receiving aid.

Attitudes towards livelihoods recovery

All households felt hopeful that they would eventually recover their livelihoods, but believed that this would take anywhere from three months to five years, with an average of 2.5 years. Most households wanted to further develop their existing livelihoods activities, in particular through working capital in the form of grants or loans: out of the 102 households surveyed, 75 gave this as their first priority, and 17 as a secondary need.

Accountability and transparency

All the LCs reported that they were ‘satisfied’ with the intervention, for various reasons: SCiM dealt directly with the entire community and consulted everyone; the community had a choice in what to purchase; the programme helped reduce food insecurity, for instance by increasing income; and it helped to ameliorate the problem of joblessness. In a few cases, LCs influenced the type of assets some households purchased. In one instance, for example, a basket-weaver was told by the LC to buy a pig even though he did not know how to raise it, and actually wanted to invest in his weaving business.

Households too said that for the most part they were satisfied with the intervention, which they thought was fair, and addressed the needs of the most vulnerable households. Households placed importance on staff taking time to explain and consult, and liked the fact that they were given a choice. Communities also thought that the LCs were an effective means of community mobilisation. There was a strong sense that fairness meant giving equal support to everyone, regardless of how well-off households were. Generally, communities felt that the better-off often acted as ‘patrons’ to poorer households, and there was a strong sense of inter-dependence.

For this reason, village-level targeting covering 100% of households was used in this emergency intervention. While 100% coverage is not normally a method advocated by SCiM, it may be appropriate in the immediate aftermath of a crisis, when targeting would have caused delays. Better-off groups would be less likely to support targeting of poorer groups until they have recovered some degree of their own assets.

Market responses to cash injections

From the interviews with traders, it seems that local markets were only marginally affected by the programme, as the cash amounts being provided were not significant compared to the size of the market overall. Although not at pre-cyclone levels, markets were very resilient. Although the prices of some goods rose steeply in the first two months after the disaster, overall household and productive supplies could be bought at or just above pre-cyclone prices. In our household surveys, half of respondents said that markets were recovering and shops reopening, but that there were still fewer market actors than before, and some items had gone up in price. For example, certain types of boats had gone up from 150,000 kyats to 280,000, nets from 30,000 to 40,000 kyats, diesel from 4,200 to 7,000 kyats per gallon, beef from 3,000 to 5,000 kyats, chickens from 3,000 to 6,000 kyats, and salt from 300 to 800 kyats. Any price increase must also take into account the rampant inflation in the country, which stood at 34% in 2007.

Some traders were able to purchase on credit, albeit only in small amounts of 200,000–300,000 kyats. As a result of unpaid debts after the cyclone, goods were sold less frequently on credit. Traders who were still selling on credit were charging between 10% and 30% interest a month on the value of the goods. Some shopkeepers mentioned difficulties in restocking as a result of problems with damaged storage space.

Conclusions

Several encouraging conclusions can be drawn from the evaluation of SCiM’s project. First, cash transfers can be a very effective and quick way to help people recover their livelihoods after disasters, whether through access to capital as a loan or grant, or through in-kind asset replacement and skills training. In Myanmar, because the majority of poor and very poor households still rely on casual labour as their predominant source of income, future livelihoods interventions should focus on the replacement or provision of livelihoods assets in order to smooth out household cash flow, as paid labour is highly seasonal. Second, the fact that none of the households asked for food aid to carry on suggests that people prefer assistance with livelihoods recovery over continued food aid. People clearly appreciate choice and an opportunity for consultation. Third, markets are resilient and can provide most of the goods people need to restore their livelihoods. Any type of cash transfer programme should be followed up with monitoring visits to see that households are using their cash for the purposes intended. Ideally, it should also be followed with other forms of support to help people utilise their grants effectively, such as business development or skills training.

Sue Mark is Emergency Livelihoods Advisor for SCiM. Her email address is siusuemark@gmail.com.