Much has been written about decentralisation and the local civil society institutions-local governments interface, although the political character of the process, particularly in developing countries, has often been underestimated. Decentralization and participation are both means of bringing a broader section of a given population into public decision-making processes - in a role of informing and/or controlling those processes (Ribot, 1999: 1). The assumption is that greater participation in public decision making is a positive good in itself, and/or that it can improve efficiency, equity, and, especially important in the context of local institutions involved in DRM, development and resource management. By bringing government decision-making closer to the citizenry, decentralisation is widely believed to increase public sector accountability and therefore effectiveness, whilst contributing to the strengthening of a genuinely people-centred type of democratic culture.
Unwittingly, local population groups themselves often do their part to further increase their risk exposure, for example, through unsafe settlements on steep slopes, unsustainable deforestation leading to soil degradation, etc. This has been the unintended outcome of a lot of spontaneous household relocation and formal resettlement programmes promoted by government, in which newcomers lack the necessary local agro-ecological and farming systems knowledge to devise risk coping strategies suited to their novel surroundings. This constitutes all the more reason to inform local population groups about the risks they are exposed to, involving them as responsible actors in disaster prevention activities - usually, everyone has something to contribute to the reduction of disaster risk hazard and should be provided with the opportunity of doing so. It is a way of increasing the self-reliance of the population at risk and the sustainability of disaster prevention measures, all of which is more easily achieved at the decentralised, local government level, and below.
As Crook and Manor (1998) argue, bringing government closer to people increases efficiency by helping to ...tap the creativity and resources of local communities.... Decentralisation is believed to increase coordination, vertical linkages (discussed in the Section 8 below) and flexibility among administrative agencies and effectiveness in development and conservation planning and implementation. Where it is real, local government bureaucrats and technocrats are in a position to invest in DRM as they have been devolved the power and provided with sufficient funds to do so. It is however often difficult to use scarce public funds for environmental conservation - which contains elements useful to natural disaster prevention - given the unattractiveness of such activities in political terms; on the other hand, it is the aftermath of catastrophic events that provides opportunities to accumulate political capital through (donor-funded) infrastructure reconstruction.
In sum, if participatory, decentralisation can increase managerial and economic efficiency by: allowing local population groups who bear the costs of resource use decisions to make those decisions, rather than leaving them in the hands of outsiders or unaccountable locals, increasing efficiency by internalising economic, social and ecological costs and benefits; reducing administrative and management transaction costs via the proximity of local participants, access to local skills and local information; and using local knowledge and aspirations in project design, implementation, management and evaluation for better matching of actions to needs (adapted from Ribot 1999). Given that natural disasters rarely hit whole countries, but, rather hazard risk often varies even from one micro-region to another, it becomes essential to use local knowledge for effective prevention measures and to adapt these to local threats and vulnerabilities. Whereas this tends to happen within the confines of DRM project frameworks, (as we will see further below) this is far from being institutionalised in the public sector.
National disaster plans may mention mitigation and preparedness, but often lack detail and dedicated resources. Social, political and macroeconomic pressures can undermine the capacity of state authorities to reduce risks. Cash-strapped central governments may simply abdicate their responsibilities, leaving disaster management to local governments and NGOs, even though they (know they) lack the skills and resources to do so. In many parts of the world, fiscal and financial decentralisation have in fact not kept up with the pace of politico-administrative decentralisation. Local governments can thus often only count on a narrow tax base and are not usually devolved sufficient central funds to be able to afford luxurious expenses such as those committed to NRM, let alone ex-ante investments in DRM, which remain difficult to justify vis-à-vis local constituencies often angry at continuing budgetary cuts in the health and education sectors, for example.
Notwithstanding this state of affairs, initiatives have been taken in many regions, although it appears to have happened mostly through externally-financed projects and programmes. Where these keep on occurring repeatedly, contributing to the creation of a handout syndrome, government post-disaster relief compensation programmes and international assistance may also act as incentives for people to locate to disaster-prone areas (Charveriat, 2000). Certain more recently instituted local government arrangements serve, amongst other objectives, to contribute to alleviate the plight befalling the coffers of the local administration: beyond their role in cross-cultural exchange, twinning programmes between municipalities in the South and the North, for example, seek to also bring together all types of resources and experiences to that effect. They are often among the first channels to be appealed to in order to mobilise supplementary external funds to deal with emergencies such as those caused by natural disasters, and, often more importantly, play an important advocacy role vis-à-vis regional and national governments as well as, sometimes, the international community.
As intermediaries with more affluent urban or industrialised rural environments, often in Europe and in the USA, migrant associations have sometimes also been carrying out comparable functions. The impact of their activities tends to be more localised, at the intra-community level, not least because of the influence of clan-based and other kinship related social networks within them. Although this is yet another area in which research data are scarce, it appears that the commitment of such associations to play a prominent role in DRM depends partly on their degree of politicisation, both vis-à-vis their host countries and those of their origin. With diasporas acting as financiers and fund-raisers for the immediate rehabilitation needs of the households to whom they are connected through familial ties, the bulk of such assistance however tends to take place at the individual level, a phenomenon facilitated by the increasing international outreach of commercial money transfer services and the concomitant diversification of transfer options.
7.1 Local Emergency Committees
In Costa Rica there exist more than 60 Local Emergency Committees, at the local administrative level of the canton, composed of the delegates of various institutions, with each member being assigned a role in case of an emergency; these bodies are integrated into Regional Committees.
The local committees aspire to be facilitators of community mobilisation and organisation. Only some few institutions are however represented permanently on the committees, which limits their possibilities of planification and action. In certain instances, activities are carried out such as the laying out the inventory of resources available to face emergencies and the establishing of several brigades (rescue, first aid, food distribution, transport, etc.). These activities lead to the drafting of Local or Regional Emergency Plans. The latter provide the population with information on where to go in case of an evacuation alarm, who will assist and be assisted, and which other activities are to be joined.
Unsurprisingly, the extent to which DRM measures are institutionalised and streamlined within local government systems depends largely on the regularity and intensity with which their constituencies keep being affected by extreme natural events. In many parts of the world, during periods of heavy rainfall for example, clogged up drains can have a dam effect preventing water from flowing freely, thus creating overflows and ultimately giving rise to flooding. In most of Latin America and the Caribbean, like in Jamaica, drains maintenance is carried out by government and local parish councils, which include the Parish Disaster Committees. These are local government bodies providing residents with evacuation procedures and are responsible for organising and directing local disaster preparedness and emergency relief operations in collaboration with other voluntary agencies and the governments emergency services (see also the box in this Section).
In coastal Asia where flood risk is severe, for example in Bangladesh and Cambodia, several projects have been built around the concept of specifically focusing on: peoples perception of flood risk; the purpose and tools of community flood risk assessment; the strategies for community organisation; and resource mobilisation and capacity building. In these cases, the rationale for doing so can be traced back to the sequencing of DRM activities, with an emphasis on local scoping studies and capacity building that are to precede community interventions. In North-West China, extension mediation groups were to be used in a similar fashion. With the exception of flood-prone Mozambique, most of the relatively few examples of local government involvement in DRM in Sub-Saharan Africa, on the other hand, are not related to rapid onset phenomena such as floods; rather, they can be found in the Sahel, where the recurrent droughts and a strong associational culture coupled with advanced decentralisation have led to some degree of success in coping with these slow onset disasters.
In Mozambique, clients of the Fondo de Crédito Communitário (FCC), which applies a village or community banking methodology, were facing unprecedented floods when a cyclone hit on February 22, 2000. The flooding, considered the worst in 50 years, caused more destruction to infrastructure than the whole civil war experienced by the country until the early 1990s. Most FCC clients were displaced for about two months and the emergency response phase took two more months after they returned to their homes. As a pilot test during the emergency, cash grants were offered directly to these households by another organisation; FCC clients had the option of using this grant to pay off their outstanding debt or restructure their loan in order to keep the grant. Of a total of 89 community banks targeted, only 3 chose to restructure. The community banks that chose to repay, paid the loans before the anticipated repayment date. Evaluations suggest that the cash grants did not have any negative impact on the credit culture of FCC clients since, first, the grants were offered by a different organisation, and second, most used the grant to repay their loans in order to obtain a new loan.
 Most national Poverty
Reduction Strategy Papers (PRSP) call for cuts in government spending; e.g. in
Mozambique, the (World Bank-led) prioritisation process has meant that DRM
investment and education have lost out to health investment, making it unlikely
that even essential repairs to dykes will be carried out before the next rainy
season, if at all. In fact, the Mozambique PRSP has a section on Reducing
vulnerability to natural disaster - but no money is allocated for this