10. Mapping Local Institutions in the Disaster Risk Mitigation Cycle

10.1 Preparedness: Insurance

In developing countries, it is difficult to establish the link between insurance and mitigation. The need for mitigation is high as many structures are completely uninsurable since they are not only located in settlements without basic services and/or in flood plains or other places with high probability of disaster occurrence. In addition, many of these structures are not built with solid materials and appropriate building standards, while their occupants often lack legal ownership title (World Bank 1999). The poor, in addition, do not have adequate financial incentives, let alone the means to take mitigation actions. Local governments, at the same time, lack the capacity to develop and enforce land use management plans and building standards to improve the conditions of these settlements.

Insurers wanting to provide insurance services to the poor face the challenge of setting up affordable rates which can also ensure the financial sustainability of the programme. In the end, even if a risk is considered insurable, it may not be profitable or sustainable, since, as Kunreuther (1998) puts it,”...it may be impossible to specify a rate for which there is sufficient demand and incoming revenue to cover the development, marketing, and claim costs of the insurance and still yield a net positive profit” (p27). This was precisely the experience of insurance companies in the USA that led them to declare flood risk as unmarketable. Without a mandatory requirement it is difficult to spread the risk among a large number of people in order to provide affordable insurance rates.

Under certain circumstances, even when risks are technically insurable, there may be alternative risk management products that are more adequate, such as savings and emergency funds. It may be feasible to provide disaster insurance services, but given high risk exposure levels, insurance premiums will most likely have to be set at a rate that only few can afford. In countries like Mexico, for instance, low coverage stems in part from the high premium prices that the insurance industry has to charge in zones that are highly prone to earthquakes (World Bank 1999). In low risk exposure areas, people do not have the incentive to buy insurance coverage, which further reduces the insurer’s scope to bring down costs to a viable level through cross-subsidisation.

SEWA (see Section 6 above) provides comprehensive coverage to its members. SEWA started its integrated insurance programme in 1992, as a collaboration between SEWA, SEWA Bank and the nationalised insurance companies. Eventually, SEWA established its own insurance company, VimoSEWA, which according to the most recent figures, has insured about 90,000 women and men in Gujarat. The current insurance programme offered by this institution consists of a group insurance package linked with other insurance companies. Specifically, the programme has three different insurance packages including life, accidental death, hospitalisation and maternity, and loss of housing and other assets. In terms of asset protection, SEWA provides coverage to members for losses related to natural disasters such as fire and flood, and man made disasters such as riots. Interestingly, the insurance scheme is linked with savings: women who want to become long-term members of the insurance scheme can deposit a certain amount in SEWA Bank and the annual premium is paid from the interest accrued from this deposit. Over the last five years, SEWA members have had to cope with one flood, two cyclones, three droughts, one epidemic and a catastrophic earthquake.

An approach seen as very promising are index and area-based contracts to insure natural disasters. Area-based index insurance is provided through contracts written against specific perils or events such as area yield loss, drought, or flood, defined and recorded at a regional level via, for example, local weather stations. The insurance is sold in standard units, with a standard contract or certificate for each unit sold (known as a Standard Unit Contract). All buyers are free to buy as many units of the insurance as they want, pay the same premium rate for a Standard Unit Contract in a given region, and receive the same indemnity if the insured event takes place. A good example of this type of insurance is area-based crop yield insurance. Usually, for example, in India, insurance is written against the average yield of a region, and a payment is made if the measured yield for the region falls below the pre-defined limit. Area-based yield insurance requires long and reliable series of area-yield data, a kind of data not usually available in many countries. Alternative indices such as rainfall and soil moisture can be used instead of historical area-yield data. The same borrowing groups used for savings and lending services can be used as conduit to sell area-based index insurance.

Disaster insurance, although not a mitigation strategy per se - as it redistributes rather than reduces losses - should promote the adoption of loss reduction measures. Pantoja (2002) finds that for the most part, MFIs have not managed to tap into the potential of insurance as an instrument to assist clients in their risk management efforts.

10.2 Mitigation: Microfinance institutions

Local MFIs can undertake a wide range of complementary activities to mitigate disaster risk, and thereby contribute to ensure that emergency responses become more community-based and sustainable (see Table 1 below). MFIs have an important role to play by promoting disaster risk and vulnerability assessments of their clients. Although further research is required, several factors seem to influence the effectiveness of MFIs during disasters. Those institutions with good leadership, sound financial management and accounting systems, and a certain level of disaster preparedness manage to respond faster and better to the disaster situation. Rapid access to cash, made available in the form of emergency funds or through efficient transfer of external funds, are particularly critical.

Having committed and easy to deploy field staff allows certain MFIs to carry out damage assessments rapidly and to monitor the situation closely. In turn, damage assessments and close monitoring of the situation enables these institutions to respond better to their clients’ needs, and the assessments provide them later on with more accurate estimates of the funds needed for the recovery process. Another critical factor influencing the relative success of MFIs’ assistance during disasters is the level of engagement with, and relative dependency on donors and international NGOs. Currently, involvement of microfinance in disaster risk management in many countries remains highly vulnerable to the ebbs and flows of donor funding. Given ongoing relationships, donors and governments have typically found it practical to channel emergency and recovery funds through MFIs. In fact, the major source of funds for the products and services offered by MFIs in post-disaster situations has been grants from donors. Setting up new MFIs as a post-disaster response, however, may not be effective because these institutions would lack experience, and knowledge of the area and of the affected households.

Most MFIs today would not consider debt forgiveness as part of their post-disaster efforts. Past experiences indicate that, even though debt forgiveness brought immediate relief to affected borrowers, it undermined years of work of the microfinance sector aimed at fighting the “handout syndrome” and creating a culture of repayment and financial discipline. At the same time, debt forgiveness may increase the losses suffered by the institutions and exacerbate their liquidity constraints, while making them more dependent on donors or government support. Although context-specific, the long-term negative consequences on the impact and sustainability of the microfinance institution may often probably be higher than the immediate benefits enjoyed by borrowers.

In Nicaragua, the ‘Asociación de Consultores para el Desarrollo de la Pequeña, Mediana y Micro-Empresa’ (ACODEP), one of the largest MFIs in the country, has been learning from the experience of hurricane Mitch in 1998 and more recent disasters. The association has developed a ‘Disaster Prevention Plan’ whose objectives are to identify, prepare for and mitigate natural and manmade disasters in order to protect the institution, its clients and staff from possible losses. The Plan is rather comprehensive, including measures to protect the institution’s staff, portfolio, facilities, equipment and information systems and records, as well as measures to better respond to the many disasters that affect Nicaragua. The Plan recognises that priority should be given to assisting clients in finding medical aid, contacting relief organisations and joining FFW programmes, but, in keeping with the sector’s orthodox ‘best practices’, it does not consider that the institution should provide relief directly.

ACODEP’s plan is interesting in that it further outlines a basic, flexible credit policy for disaster emergency and recovery, and the creation of a disaster loan fund, to help the association prepare for possible cash flow demands and control credit and liquidity risks. Specifically, the plan establishes that the institution will, inter alia, stop collecting payments during the emergency period; allow clients to withdraw their deposits (which are normally used as collateral); stop lending (short-term loans of 1 or 2 months, with special interest rates, would be granted in cases of severe emergencies for household needs such as food or medicines); and, on the basis of a field damage assessment, prepare loan restructuring and refinancing plans, considering two types of situations: restructuring loans where clients lose their housing but productive assets are not affected and/or they are severely injured, and refinancing loans when productive assets are lost but clients escaped the disaster unharmed (Pantoja 2002). Most disaster or emergency loan funds have at least two levels of terms and conditions, as money needs to be transferred from them to MFIs and from there to clients.

In Bangladesh, the Grameen Bank has set up a comprehensive system to mitigate possible liquidity shortages after a disaster, based on three mechanisms operating at different levels: the group level, the centre level and the institutional level. Each lending group has to create an emergency fund towards which each member pays 5 percent of each loan; in each of the 65,000 centres, borrowers have to contribute approximately 25 percent of the total interest they owe into a centre disaster fund; and the Grameen Bank keeps US $100 m as a disaster fund.

Table 1

In the same country and in a manner similar to ACODEP, although on a smaller scale, the Association for Social Advancement (ASA) has established a permanent disaster loan product to offer to their clients after a disaster: loans range from 500 to 1,000 taka each, are interest free, and must be paid back within two years through 100 equal weekly instalments (Rahman 1999).

Faced by natural disasters, many MFIs have consistently managed to maintain discipline in their existing projects and sometimes even been able to use these events as opportunities to strengthen the sector. In several cases, they have been found to come together in an informal manner to avoid the impact of adverse decisions such as government directives to forgive debts - a relief strategy often resorted to in the past. Thus, “their efforts would be supported greatly if donors and governments agree on disaster response and recovery policies for the microfinance sector before a disaster occurs” (Pantoja, 2002: 31). Vulnerabilities will change over time as MFIs evolve and expand, and their portfolio changes.

Governments and donors have an important role to play in promoting the adoption of DRM strategies in the microfinance sector, and evaluate their results, in order to maintain appropriate policies and procedures. MFIs can carry out important functions in preparedness, reduction or mitigation and risk transfer, and response (coping) and recovery (see Table 1 and the next Section below). Unfortunately, they have been doing this much less than would be desirable, so that these examples are drawn from relatively few practical experiences.

Similar to the rescheduling of compulsory savings, loan rescheduling may help clients and protect the MFI by allowing clients to repay loans in a flexible manner. By giving affected clients the option to delay repayments on their loans for a specified time, MFIs can counteract the probability of defaults and reduce financial losses (Nagarajan and Brown 2000). Empirical evidence indicates that disaster-stricken borrowers do not necessarily insist on debt forgiveness, and are willing to accept assistance to improve their liquidity through, for instance, cash or in-kind relief loans, and access to savings. Emergency loans might be a good mechanism to help affected households, and other demand-based financial services, such as channelling remittances, which can be offered to everybody across the affected area and which would ease cash flow problems of clients and non-clients.

The provision of technical supervision is an unavoidable requirement of a housing programme if one of the objectives is to improve building standards and practices that take disaster risk reduction into consideration. Experienced MFIs do not recommend using the solidarity group lending methodology for home improvement loans.