The earthquake in Bam, Iran, in December 2003 killed more than 26,000 people, injured 30,000 and left 75,600 homeless. The earthquake completely devastated the city: 85% of buildings and infrastructure was damaged or destroyed, largely because most buildings were built according to traditional methods, using sun-dried bricks. The UN estimated that 200,000 people were affected by loss of economic activity or damage to property in Bam and surrounding villages.
The cash programme in Bam: working through the welfare system
In response to the earthquake, the British Red Cross, in conjunction with the Iranian Red Crescent, implemented a cash programme in April 2004. The programme was designed to provide financial support to vulnerable people affected by the earthquake. The amount distributed was meant to allow families moving from tents to temporary/transitional houses to purchase household items, such as refrigerators, air conditioning systems (much-needed in Bam) and furniture. The programme targeted female-headed households, orphans, the disabled and the elderly. A total of 4,551 households received a one-off distribution of cash estimated at £185. Because the banking system was reactivated soon after the earthquake, distributing cash through individual bank accounts was considered to be the most appropriate mechanism. A voucher scheme was ruled out because markets were not strong enough (most commercial activity involved small-scale operations out of ship containers). A number of families interviewed during the evaluation of the programme in July 2006 stated that delivering the cash directly would have been easier for them, but would also have created risks as their prefabricated houses often lack proper locks and are generally insecure.
The vulnerable groups identified were those supported by the government Welfare Organisation (WO). The decision was made to work in close collaboration with the WO as the most effective way of reaching the most vulnerable people quickly. An independent household identification process within a disrupted, densely populated urban environment would have taken too long and would have required human resources that were unavailable at the time. It would also have involved a disproportionate cost compared with the amount being distributed to households. The WO has a responsibility to identify vulnerable people, and was in the process of rebuilding and updating its caseload at the time. The cash grant, which was significantly higher than the monthly benefit distributed through the welfare system, was to provide recipients with relief in addition to the in-kind, and some cash, distributions being provided by the government.
The WO, like most organisations in Bam, came under intense strain following the earthquake. A number of staff died in the disaster, the WOs office building was destroyed (the organisation was still operating out of temporary shelters at the time of the evaluation), and most files and computers were lost. The WO thus faced significant challenges in rebuilding and updating its caseload. At the same time, the WOs workload had significantly increased due to the additional people rendered vulnerable during the earthquake the caseload was roughly 1,500 people before the earthquake, and is over 8,000 today.
For the cash programme, the WO was to provide a list of beneficiaries, with basic information on eligibility and status, and bank account details. It emerged that this information was not available in a single database, and when compiled there were mismatches between the information provided by the WO and that furnished by the bank, which had to be resolved. This resulted in significant delays in implementing the programme, with the first transfer happening only in May 2005 (for 2,861 households whose details had been verified by that date). However, the evaluation found that the delay was not a significant issue for the families involved. Although the aim of bringing additional support when moving from tents to temporary/transitional houses was not met, the cash still provided a necessary relief resource to meet these households ongoing needs. However, one lesson from this programme is that the strain of a disaster on a local governments offices should not be underestimated. More support to the WO, or possibly a more direct role in establishing the list of recipients, would in retrospect have speeded up the programme. That said, the programme benefited from the experience of the WOs relief workers, who had direct contact with the families, and a very good knowledge of local communities.
Targeting the most vulnerable
A key question for the evaluation was whether the most vulnerable were reached, and who might have been excluded. Iranian society accepts the need for targeting in certain circumstances (for instance, the zakat principle one of the five basic precepts of Islam is a moral obligation for Muslims to care for the poor, hungry, widows and orphans, and is collected as a regular tax), but in emergency situations the view prevails that everyone deserves the same support. There were also legitimate concerns within the Iranian Red Crescent, which is mandated by the Iranian government to respond to disasters, that introducing targeting while it was undertaking a blanket distribution of relief items and cash on behalf of the government would create confusion, and could exacerbate tensions between residents of Bam and those who had moved to the city after the earthquake and were also claiming relief aid. There was also a practical consideration: the Iranian Red Crescent, which played a crucial role in the relief operation, did not have the time or the capacity to conduct a thorough vulnerability assessment.
The criteria of vulnerability used by the WO are based on belonging to a particular group (such as female-headed households, orphans and the disabled). Even if further individual interviews by social workers determine whether a family is actually entitled to receive support, other factors of vulnerability, such as the level of income or whether the family is landless, are not considered. A household economic survey, using economic security criteria, would have provided more accurate information on vulnerabilities that could be appropriately addressed by a cash intervention, but would have taken precious time. There is a fine balance between the accuracy of targeting and the time and resources it takes to collect the information. In this case the balance was appropriate, though it is likely that a small proportion of recipients were not strictly in need of additional relief cash, perhaps because they had alternative sources of income or their vulnerability required protection interventions rather than additional cash. This small inclusion error is not of significant concern to a humanitarian agency with a relief objective. The evaluation found that the targeting was effective in reaching significantly vulnerable people.
Of more concern is the exclusion error of the programmes chosen targeting mechanism. It appears that some vulnerable families who would have been eligible for cash support were unintentionally excluded because the WOs work to rebuild and update its overall caseload had not been completed by the time BRCS/IRCS required the lists. Once the list of recipients had been established, lack of funds meant that newly registered families would not be included, even though, by the time the cash was eventually transferred, significant additional numbers on the WO caseload met the criteria. This is a dilemma with cash programmes that are planned as one-off payments. In an ideal world, sufficient funds should be held in reserve so that all households meeting the same criteria can be included once the lists were updated. At a minimum, for reasons of expediency, a transparent cut-off date should have been communicated to the affected population, after which no further names would be added. Instead, as duplications were identified in the lists provided WO staff put others forward. This did not adequately address the exclusion issue.
Lessons from cash programming in urban settings
Although the evaluation did not specifically consider the urban nature of this response, some reflections are possible. Rural and urban areas were affected differently by the earthquake. While most of Bam city was destroyed, there was less damage in surrounding villages. While the rural population suffered a smaller number of deaths and injured, even areas with no direct effects were still indirectly affected by the devastation in the city: their livelihood largely depends on Bam citys economic activities, which provide an estimated 250 villages nearby with their main source of employment and trade.
Reconstruction has progressed differently in rural and urban areas. It has been much faster in villages, partly because there was less damage and the government took full responsibility for arranging and managing construction contractors, but also because the process has been much more complex in the city. Leaving aside the risks of aftershocks and the huge task of clearing the rubble, complex land tenure issues and reaching agreement over an improved city master plan contributed to delays.
The cash programme mainly supported inhabitants of the city; only a small proportion of recipients lived in the villages. In its design, the programme has not distinguished between urban and rural areas. Should the cash programme have been planned and implemented in a different way for these two different settings?
Because the groups being targeted were identified by their status, there was no major difference in terms of vulnerability according to whether people lived in rural or urban areas. Their needs seem to be comparable. Things would have been different for a cash programme aiming to restore livelihoods, where the amount of cash provided is determined by the type of livelihood activity employed.
Another possible distinction relates to the banking system. As there are few banking facilities in the villages around Bam, using banks to transfer cash to this rural population may have created an additional burden by making recipients go to town to collect money. In fact, most families interviewed argued that this was not problematic, as they have to go to the city anyway to collect their monthly benefit from the WO and to go shopping. Had the recipients not also been receiving support from the welfare system, a different cash disbursement mechanism may have been more appropriate.
As mentioned above, another set of issues to consider for cash programmes in urban settings relates to targeting. Approaches typically used in rural contexts may be logistically difficult in densely populated cities; community-based targeting and verification may be difficult, for example, when urban communities know little about each other. Working through the welfare system in this context seems an appropriate approach. Had the surrounding villages been the particular focus of the programme, a different approach would have been required, as the WO does not have strong outreach to these areas.
Despite the delays in distributing the cash, the programme has provided much-needed support to the families included in the programme. Working through the welfare system was for many reasons an appropriate choice in this context, although it did lead to a significant exclusion error. However, more support to the WO would probably have helped address the most significant delays, and the alternative of carrying out an independent beneficiary identification process would have entailed disproportionate costs. The delay was not perceived as hugely problematic by most families interviewed, although many institutional informants felt that, as a relief intervention, the cash should have been distributed earlier.
Not surprisingly, most families were very clear that they much preferred receiving cash than in-kind items. Whereas most families used part (or all) of the money for daily items and needs, many also saved some money for the future (particularly when the cash was provided for orphans, who had often been registered as part of the WOs protection mandate), or for school fees and household equipment. Because of the delays in the programme, fewer families purchased household items than initially anticipated, as they had already received such items from other sources. Recipients appreciated the fact that they had control over the use of cash, a significant factor in maintaining their dignity, and the range of uses to which the cash was put suggests that an in-kind distribution would not have met families diverse needs.
Mansooreh Bagheri is Programme Coordinator for International Operations, Iranian Red Crescent Society. Vendela Fortune is Programme Development Adviser, British Red Cross Society. Charles-Antoine Hofmann is Humanitarian Policy Adviser, British Red Cross Society.