65786 Natural Disaster Response Lessons from Evaluations of the World Bank and Others 16 Natural Disaster Response Lessons from Evaluations of the World Bank and Others David Todd and Hazel Todd Evaluation Brief 16 http://ieg.worldbankgroup.org © 2011 Independent Evaluation Group Communications, Learning, and Strategy The World Bank 1818 H Street, NW Washington, DC 20433 http://ieg.worldbankgroup.org Email: ieg@worldbank.org Telephone: 202-458-4487 All rights reserved This Evaluation Brief is a product of the staff of the Independent Evaluation Group (IEG) of the World Bank. The find- ings, interpretations, and conclusions expressed here do not necessarily reflect the views of the Executive Directors of the World Bank or the governments they represent. IEG does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank or IEG concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. IEG encourages the dissemination of its work and will normally grant permission to reproduce portions of the work promptly. For permission to photocopy or reprint any part of this work, please send a request with complete information to ieg@worldbank.org. ISBN-13: 978-1-60244-202-3 ISBN-10: 1-60244-202-9 Contents Abbreviations v Summary vii 1. Introduction 1 2. The Phases of the Disaster Management Cycle 3 3. Lessons for the Pre-Disaster Phase(Phase 1) 7 4. Lessons for the Disaster Response Phase (Phase 2) 11 5. Lessons for the Post-Disaster Phase (Phase 3) 19 6. A Lesson Concerning Slow-Onset Disasters 27 7. Applying the Lessons 29 Bibliography 31 Abbreviations ERL Emergency recovery loan GRM Grievance Redressal Mechanism ICR Implementation Completion and Results Report IDA International Development Association IDB Inter American Development Bank IEG Independent Evaluation Group IFC International Finance Corporation NGO Nongovernment organization M&E Monitoring and evaluation v Summary This document aims to provide a concise collec- • Capacity development in disaster manage- tion of those good practices and challenges ment before a disaster strikes can reduce the that have recently been identified by evalua- burden on disaster response support, make it tive work on natural disaster response of the more effective, and increase national owner- World Bank, the Independent Evaluation Group ship of the response process. (IEG), and other major actors. It is hoped that • A balanced approach to disaster manage- the lessons identified can be used by World ment should relate it to broader national ap- Bank staff engaged in work related to natural proaches to sustainable development and disasters, to gain insights into approaches that poverty reduction. may promote or hinder effective assistance to partner countries. Lessons for the Disaster Response Phase Although great international attention is focused Evaluations have provided many lessons for on the occurrence of disasters and the short- this phase. However, these generally also carry term response to them, this phase is only part over into the post-disaster phase, and even of the overall disaster management cycle. This further into the next pre-disaster phase. Some cycle includes three main phases—pre-disaster, of the lessons may partially contradict each disaster response, and post-disaster—each of other (for example, the value of local partici- which has an appropriate range of activities. pation may cut across the benefits of speedy These phases do not have clear boundaries, but action), which means that trade-offs must be overlap chronologically, as well as in terms of developed for each specific situation. Main the ongoing activities. In placing the lessons emerging findings are presented below: concerning natural disaster response projects within the broader context of the disaster manage- • Keep project design as simple and realistic as ment cycle, it becomes clear that responses can possible within what is likely to be a complex be simpler and more effective if activities of the implementation environment. pre-disaster phase have already been appropri- • Base project deadlines on a specific analysis of ately conducted. Nevertheless, because this is capacity in the situation of post-disaster disrup- often not the case, the lessons presented take tion, rather than on the “normal� situation. account of the possibility that existing institutions • Streamline procedures as far as possible to dealing with disasters may not be strong. meet the need for quick delivery. • Although essential, the rapid development Lessons for the Pre-Disaster Phase and processing of disaster-related interven- tions present ownership challenges. Many lessons for the pre-disaster phase of the • Natural disaster responses need to include a disaster management cycle actually appear in broad range of stakeholders, especially the evaluations of disaster response activities. This poorest and most vulnerable, as well as the is because pre-disaster activities have largely private sector. been catalyzed by such situations. Two basic • Disaster response situations often present lessons are apparent for this phase: the best opportunities to develop disaster vii evaluation brief 16 preparedness (pre-disaster) activities, because and activities are vital in post-disaster situa- of the temporarily high profile of disasters. tions, particularly when major infrastructure • Building such disaster preparedness elements activities are implemented by multiple stake- during the response stage may require a longer holders. period of implementation, initially focusing on • Because disaster-related projects face un- rehabilitation and later on mitigation. usual implementation barriers associated • Information, communication, and data manage- with weakened institutions that are expend- ment systems, which are vital in any disaster- ing substantial funds quickly, strong monitor- related situation, have often proved weak. ing and evaluation systems are even more • Agencies financing natural disaster response important than normal. projects have reported significant challenges in designing an instrument that provides re- The Specifics of Slow-Onset Disasters sources quickly enough or for long enough to achieve their objectives. Slow-onset disasters, such as droughts, are often long-lasting and recurring situations. To Lessons for the Post-Disaster Phase deal with this difficult combination, there is a strong need for collaboration and coordination Evaluations of post-disaster phase activities between stakeholders involved in both humani- have similarly generated lessons, many of which tarian and developmental activities. However, anticipate the next round of pre-disaster activi- this collaborative relationship has often proved ties, as the cycle re-commences: difficult, and cooperation between the two kinds of institutions is widely reported to be • Post-disaster recovery requires strong institu- ineffective. tions, which can be achieved either through developing the capacity of existing bodies or Applying the Lessons by creating new ones. The latter usually need specific mandates and time boundaries to be Since the publication of IEG’s 2006 evaluation effective. Hazards of Nature, lessons have continued to • Community-driven approaches to recovery emerge concerning natural disaster response and reconstruction can help build local ca- activities. These lessons have been organized pacities for future project identification, plan- here according to the phase of the disaster ning, implementation, and ex post operation management cycle in which they first occur. and maintenance. • Housing reconstruction programs that give One overriding lesson concerns the advantages of maximum responsibility to homeowners have reducing risks and increasing preparedness within been found more effective than contractor- the national development strategies of countries led systems. However, these programs need that are prone to disasters. Measures taken to reach specific measures to ensure that vulnerable these objectives are believed to deliver substantial people can participate. gains in the effectiveness of the response when • Although speed is important in designing and disasters occur. However, formal evaluations implementing reconstruction programs, this of the impacts delivered by risk reduction and should not be at the expense of accountability preparedness so far are largely absent. and transparency. • Reconstruction programs should ensure that It is also clear that every natural disaster has effective grievance and complaint mecha- unique characteristics, so that the lessons nisms are in place for those who may feel presented here need to be carefully assessed excluded. for their applicability in each case. They provide • Flexibility (especially for procurement proce- a set of guidelines that can be examined for dures) and the ability to revise project targets their relevance to specific situations, with the viii natural Disaster response: lessons from evaluations of the WorlD bank anD others objective of making interventions more effective There is also a distinctive set of lessons concern- and efficient for affected countries, as well as for ing the processes through which disaster donors’ assistance programs. response can be prepared and planned for in advance. These highlight the importance, Some of the lessons derived from project and within the overall disaster management cycle, program experience in the area of natural of disaster risk reduction and the need to raise disaster response would apply to projects in its profile in both affected countries and interna- any field, but are even more important in the tional organizations. Risk reduction should have case of disaster projects, because of the social, a central role in any disaster-prone country’s economic, institutional, and governmental overall sustainable development strategy and disruption in which they operate. should therefore form an integral part of the donors’ programs and country strategies. ix 1. Introduction T his brief aims to provide a concise collection of those good practices and challenges that have recently been identified by evaluative work on natural disaster responses of the World Bank, the Independent Evaluation Group (IEG), and other major actors. It is hoped that the les- sons identified can be used by World Bank staff engaged in work related to natural disaster management to gain insights into approaches that may promote or hinder effective assistance to partner countries. Its main data sources include Implementation This report first presents an overview of the Completion and Results Reports (ICRs); self-evalua- key phases of the disaster management cycle, tions by the World Bank of projects that closed which have been identified by academics and since 2006; IEG’s independent field-based evalua- practitioners in the field. This is important to tions of a subset of these projects (Project Perfor- show the interrelationship between the differ- mance Assessment Reports); IEG evaluative notes ent phases associated with disasters and the on specific natural disasters in Haiti, Pakistan, and types of assistance that are most effective in West Africa; World Bank publications covering preparing for and responding to such events. It natural disasters (mostly published since 2006); then outlines the different types of lessons that and reviews and evaluations of natural disaster have been found to apply to each phase before activities by a variety of international and national drawing some brief conclusions. bodies active in the field. 1 2. The Phases of the Disaster Management Cycle A lthough great international attention is focused on the occurrence of disaster events and the short-term response to them, this phase is only a part of the overall disaster management cycle. As shown in Figure 1, the cycle includes three main phases; for each phase, a range of activities is appropriate.1 These phases do not have clear boundaries, risks throughout a society to avoid (prevent) but overlap chronologically, as well as in or limit (mitigate and prepare for) the adverse terms of the ongoing activities. In placing the impacts of hazards, within the broad context of lessons concerning natural disaster response sustainable development. projects within the broader context of the disaster management cycle, it becomes clear The disaster risk reduction framework that responses can be simpler and more encompasses the following fields of action (UN effective if activities of the pre-disaster phase ISDR 2002, p. 23): have already been appropriately conducted. Nevertheless, because this is often not the • Risk awareness and assessment, including case, the lessons presented here take into hazard analysis and vulnerability/capacity account the possibility that existing institu- analysis tions dealing with disasters may not be strong • Knowledge development, including educa- at the time the disasters occur. tion, training, research, and information • Public commitment and institutional frame- Phase 1: Pre-Disaster works, including organizational, policy, leg- islation, and community action The pre-disaster phase is, in many ways, the most • Application of measures, including environ- important. With rare exceptions, the occurrence mental management, land-use and urban of disasters is in principle predictable, although planning, protection of critical facilities, appli- not their precise location, timing, or severity. cation of science and technology, partnership For this reason, a variety of measures should and networking, and financial instruments be taken in advance, notably in the areas of • Early warning systems, including forecast- risk assessment, mitigation/prevention, disaster ing, dissemination of warnings, preparedness preparedness, and risk reduction.2 There is a measures, and reaction capacities. close link between disaster risk reduction and preparedness. Risk reduction activities within Mitigation can include structural and nonstruc- the pre-disaster phase often focus particularly tural measures undertaken to limit the adverse at the community level. They fall within the impact of natural hazards, environmen- conceptual framework of elements considered tal degradation, and technological hazards. likely to minimize vulnerabilities and disaster Preparedness helps the community cope better 3 evaluation brief 16 Figure 1: Phases of the Disaster Management Cycle Phase 1: Pre-Disaster Disaster Event Early warning Risk Assessment Pre-Disaster Phase and Mitigation/ Risk assessment evacuation disaster risk Mitigation reduction activites Preparedness activities Post-Disaster—B Response Phase Social and economic Disaster response Phase 2: recovery and Initial damage Response rehabilitation assessment Ongoing development Immediate assistance to strategies and affected community activities Post-Disaster—A Media attention Continued response and assistance to affected community/international agencies Infrastructure restoration Phase 3: Post-Disaster Source: Based on work of Ian Davis, Cranfield University, Bedford, United Kingdom. should another disaster develop. It includes latter of which attempts to begin to re-establish activities and measures taken in advance to functionality of systems and infrastructure. ensure effective response to the impact of hazards. It may include leadership training and • Once a disaster has taken place, the first con- strengthening community participation; plans cern is effective relief—helping those affected for the issuance of timely and effective early to recover from the immediate effects of the warnings; and measures for the temporary disaster. Such relief work includes provid- evacuation of people and property from threat- ing food, clothing, shelter, and medical care ened locations, such as communities living near to victims. For rapid onset disasters, such as a possible source of flooding. earthquakes, this phase may last for weeks or a few months. For slow onset disasters, such as Phase 2: Response droughts, it may last months or even years. • Medium-term response takes the first steps The response phase begins immediately after a toward recovery by assessing damage to in- disaster strikes and encompasses both immediate frastructure, communities, institutions, indus- response (relief) and medium-term response, the try, and business and by planning measures 4 natural Disaster response: lessons from evaluations of the WorlD bank anD others necessary to restore these to previous levels • Restoring the basic services needed to enable or better. life to move back toward “normalcy.� • External support, such as loans to govern- There is an overlap between the response ments, technical assistance, resources for and the post-disaster phases, and projects or farmers, and help for businesses to restart. programs may span both. • Rebuilding homes and industry, which is linked to restoration of social and economic Phase 3: Post-Disaster development. It is important at this stage to design stronger buildings that are able to The post-disaster phase includes activities in the withstand future disasters. fields of recovery, rehabilitation, and reconstruc- • Activities focused on enabling communities tion. It also affords an opportunity to develop to protect themselves. Such measures need to disaster risk reduction measures, which can be be particularly available to those at greatest applied during the next pre-disaster phase (that risk—the poorest and most vulnerable in the is, Phase 1). This phase includes the following: community. • Decisions and actions taken after a disaster Notes with a view to restoring or improving the 1. The authors prefer to call this the disaster manage- pre-disaster living conditions of the stricken ment cycle, because it includes the response as well community, while encouraging and facilitat- as risk management. ing necessary adjustments to reduce disaster 2. These measures have been embodied in the Hyogo risk. Framework for Action 2005–2015 (UN ISDR 2005). 5 3. Lessons for the Pre-Disaster Phase (Phase 1) T he pre-disaster phase is, in many ways, the most important. This is because disasters tend to be concentrated in certain regions or countries, although their precise location, timing, or severity is not predictable. For this reason, a variety of measures should be taken in ad- vance, notably in the areas of risk assessment, mitigation/prevention, disaster preparedness, and risk reduction.1 Capacity building and training in all aspects Because national governments have the ultimate of disaster management before a disaster responsibility for leading responses to natural strikes (Phase 1) can reduce the burden on disasters, building national and local capacity is disaster response support (Phase 2), make critical to promoting future effectiveness in this it more effective, and increase national area. A multi-stakeholder review of responses ownership of the response process. to the Asian tsunami confirmed the importance of raising local capacity and engagement in the An important role for international partners is to entire disaster management cycle and provided help strengthen the capacity of in-country institu- specific lessons (Box 1). tions (including civil society actors) mandated to lead disaster preparedness and response, A number of Bank project self-evaluations (ICRs) particularly at the local and community level, have commented on difficulties experienced in where first responses are made. adopting such a proactive approach to reduce Box 1: Specific Lessons from Responses to the Asian Tsunami • Disaster preparedness has been mainly focused on planning for have been developed, unless these lead to practical action emergency response, with a focus on tsunami-like disasters. So aimed at mitigation measures, people will lose interest in keep- far, little attention or investment has gone into early warning, ing them updated. Already in Sri Lanka, communities have preventive, and mitigation measures or to recurring disasters complained that they have identified local hazards that ac- such as floods and droughts. centuate flooding, but the government and nongovernmental • Effective disaster risk reduction interventions have the potential organizations have not made enough resources available for to strengthen grassroots institutions at the local and district the communities to take corrective actions. levels, as well as to strengthen the interface between grass- roots community organizations and local authorities. Thus the • Interventions that are based on strong partnership and links interventions can make local governance more inclusive and with local organizations, including in the private sector (banks/ participatory. financial and insurance companies), are far more likely to suc- • Although village-level hazard maps and preparedness plans ceed than one-off asset distributions. Source: DEC 2010. 7 evaluation brief 16 the effects of future disasters, even though there tion techniques. Again, although these measures is widespread agreement that this is essential. were designed as part of a disaster response For example, in the case of the Mexico Disaster and recovery program (Phase 2), they actually Risk Management Project (World Bank 2005b), promoted better standards of housing that would even though the country had been repeatedly be more resistant to any future disaster events, affected by disasters and mitigation measures conceptually part of the pre-disaster phase. The could have offered major benefits, such Sri Lanka training not only built the capacity of measures had not been incorporated into the homeowners and artisans to a level that enabled country’s planning for sustainable development. the predominantly homeowner-driven housing To help strengthen national-level support for a reconstruction program to achieve its targets, but proactive approach to disaster preparedness, it also raised the capacity of the local construc- the self-evaluation suggests increased attention tion industry through ensuring additional skilled to raising the capacity both of the development labor. The project also developed operational agency staff and of borrowers. manuals, which the government officially adopted. These clarified the roles and respon- Although there have been such cases of sibilities of the multiple players and partners weak approaches, the Bank has elsewhere involved in the program and facilitated smooth implemented varied and innovative capacity implementation. Some of the good practices in building and training activities. Even though the preparation of the Operational Manuals are they are conceptually part of the pre-disaster shown in Box 2. phase, these have often been introduced as part of disaster response packages (that is, in Phase A balanced approach to disaster manage- 2) or as part of reconstruction programs (Phase ment is needed that can relate this area of 3). The sustainability of gains made through this activity to both sustainable development type of activity can best be assured by feeding and poverty reduction. Evaluations by the them into future Phase 1 programs. World Bank and the Inter-American Develop- ment Bank (IDB) support the importance of The Uruguay Foot and Mouth Disease Emergency including such an approach in the formal Recovery Project (World Bank 2010d) provided support strategy for the country. strong and continuous education and training for public and private veterinary services, as well as The distribution of natural disasters is broadly for farmers and the general public to update the predictable, with many countries facing recurring evolving sanitary threats. In the Ethiopia Produc- events. In such countries, it is important to clarify tive Safety Nets Project (IEG 2011b), it was discov- how linkages will be made among prevention, ered that the delivery of training programs for preparedness, response, and recovery, so that existing staff should take into consideration the the assistance provided will bring benefits in typically high turnover of rural civil servants. The the future, as well as respond to current needs. Iran Bam Earthquake Emergency Reconstruction If attention is not paid to supporting overall Project (World Bank 2010a) called for specific natural disaster management processes within additional capacity building. It discovered a sustainable development framework, it is the particular importance of the Bank rapidly likely that funds will be repeatedly required understanding and serving the client’s needs to address the same recovery needs. This calls when there has been little previous interaction by for an enhanced emphasis within operations in preparing in advance a capacity-building program such countries on pre-disaster activities (Phase on the Bank’s processes and procedures. 1) to help reduce the need for repetitive disaster response (Phase 2) activities. The Sri Lanka Tsunami Emergency Recovery p t a a t Program rovided echnical ssistance nd raining The Algeria Natural Hazard Vulnerability for artisans and homeowners in safe construc- Reduction Project encountered such a situation, 8 natural Disaster response: lessons from evaluations of the WorlD bank anD others Box 2: Good Practices in the Preparation of Operational Manuals in Sri Lanka’s Post-Disaster Situation • Planning for participatory and inclusive preparation of the • A binding implementation framework that mandates official manuals, in the light of international practices, tailored to local adoption of the manual and monitoring mechanisms/periodic needs. audits for ensuring compliance. • Clear advance clarification of the roles and responsibilities of • The manual remains “alive�—in terms of being responsive to multiple players and partners, including government institutions, emerging requirements and localized solutions—within the civil society, and beneficiaries. Although an operational manual is confines of the program policy framework. finalized over a period of time, a set of immediate operating instruc- tions can help in the early stages of program implementation. Source: World Bank 2009g. in which inadequate risk reduction and mitiga- strongly emphasized in several documents, tion measures made disaster response a much emergency and post-disaster activities still larger task than necessary (World Bank 2007a). receive much broader attention. The ICR reports that the impact of the torren- • A mismatch between IDB programs and ac- tial rains of November 2001 on Algeria’s urban tivities and actual country priorities, incentive areas would have been much less dramatic had structures, and implementation capacities. the government taken measures to reduce the As a result, countries are not able to put into risks of flooding and implemented its existing practice the new approaches developed with regulations concerning urban development. assistance from support programs. With Bank support, the government had already • An imbalance of countries’ priorities con- implemented appropriate risk reduction measures centrating on sustainable development and in the greater Algiers metropolitan area. on poverty reduction and neglecting disaster prevention, even in severely prone regions. However, these measures were also needed in other urban areas, for which development plans The IDB evaluation concludes that the culture should have included appropriate measures for of rehabilitation and reconstruction that has preventing severe damage from natural disasters, prevailed in the countries affected by natural including earthquakes and floods. Furthermore, disasters—as well as in the institutions annual budgets of agencies in charge of risk that finance those activities and in bilateral prevention and management should have included aid—should shift to facilitate preparation adequate and well-identified amounts for natural and execution of prevention and mitigation disaster prevention. The government could have programs. built on the achievements of Bank-supported projects to raise the awareness of local authorities IEG’s note on the response to the Pakistan floods and technicians regarding the benefits of preven- (IEG 2010b) also points out, for example, that tive measures and to encourage their inclusion in some flood programs have focused too heavily local development plans and activities. on rebuilding infrastructure and not enough on better adaptation and preparedness for the The failure to support a balanced approach future through complementary investments, to disaster management has also been such as flood management, cropping pattern flagged as an important issue in an evalua- adjustments, rural finance, enhancing capaci- tion conducted by the IDB (IDB 2004). ties of water users groups, and early warning The evaluation found the following: systems. There are trade-offs between the need for urgency, which might be met by simply • An imbalance of instruments across the di- replacing what was lost, and the desire to “build saster risk cycle. Even though prevention is back better,� which may take longer but has 9 evaluation brief 16 greater disaster risk reduction, mitigation, and of damage from future flooding. One of its aims long-term development results. is to improve ecosystem resilience through better rangeland management and increased carbon Several projects and programs have taken the sequestration2 (above and below ground) using opportunity to improve on prior physical as a range of technologies that enrich soils with well as organization structures, systems, and organic carbon and improve biomass production processes. The IEG note indicates, for example, and vegetation cover. that in Pakistan, World Bank flood response projects have also focused on reducing vulner- Notes ability and building farm resilience to reduce the 1. These measures have been embodied in the Hyogo effects of subsequent flooding, an approach that Framework for Action 2005–2015 (UN ISDR 2005). has increasingly appeared in national strategies. 2. Carbon sequestration describes long-term storage Similarly, a current International Development of carbon dioxide or other forms of carbon to Association (IDA) Mali Agricultural Productivity either mitigate or defer global warming. It has been Project (World Bank 2010b) has a subcompo- proposed as a way to slow the atmospheric and nent to modernize farming systems and supply marine accumulation of greenhouse gases, which are chains that is designed to contribute to reduction released by burning fossil fuels. 10 4. Lessons for the Disaster Response Phase (Phase 2) T he response phase begins immediately after a disaster strikes and encompasses both immediate response (relief) and medium-term response, which attempts to begin to re-establish functionality of systems and infrastructure. The World Bank is not normally a major stake- holder in relief activities, which are mainly mandated to be implemented by the United Nations and specialist nongovernmental organizations (NGOs), such as the International Federation of Red Cross and Crescent Societies. Medium-term response takes the first steps A number of important characteristics have been toward recovery by assessing damage to evaluated as contributing to effective natural infrastructure, communities, institutions, disaster response support. These encompass industry, and business. This phase also covers speed, inclusion, ownership, transparency, planning measures necessary to restore these accountability, and flexibility. However, it is areas to previous or better levels. Chronologi- also clear that these characteristics may to some cally these medium-term response activities often extent challenge each other so that, for example, overlap with those of the immediate response the need for inclusion and ownership may make (relief) period. This overlap has posed serious it far more difficult to carry out activities with problems because of the different mandates and speed. In the context of specific field situations, emphases of the two sets of activities and of the the ability to prioritize and establish trade-offs national and international institutions that are between the various desirable program charac- primarily responsible for them. teristics therefore becomes essential. This issue provides an important caveat to the following In principle, national ownership of all disaster lesson, which has emerged from several evalua- response activities is a goal to which all tive sources. stakeholders subscribe. In practice, however, governments weakened by disaster events may The design of natural disaster response be overwhelmed by the large scale of external activities should be as simple and realistic interventions, leading to a great reduction in the as possible. extent to which they can actually control and coordinate them. This lesson needs to be placed within the context of other lessons here, even if it may not immedi- Many World Bank activities are conceived, ately appear to be compatible with them. We prepared, and planned during the disaster can therefore reframe the lesson to indicate that response phase. They are therefore covered project design should be as simple as possible, in this section of the report, in terms of their without reducing the importance of achiev- characteristics and lessons. However, they may ing the full participation of local communities be largely implemented during the post-disaster and other intended beneficiaries (summarized phase (Phase 3). from IEG 2010c) or of taking into account local 11 evaluation brief 16 implementation capacity. Because capacity to targets and deadlines. For example, the Sri Lanka use aid effectively is often limited if institutions Tsunami Emergency Recovery Program was are weak and governance is poor, the focus rapidly processed under the Bank’s emergency from the beginning also needs to be on the guidelines. Although its targets were largely development of capacity and improvement of achieved, the government machinery was governance, not merely the reconstruction of stretched to the limit, and time extensions were physical infrastructure. One aspect of simplicity requested but not approved. Although there is that disaster response projects should limit the appear to be various reasons for project time number of implementing agencies and sectors overruns, it is clear that an accurate assessment involved. It is also appropriate to minimize the of the effects of the disaster on the capacities of conditions placed on the lending. Implemen- government and implementing agencies is an tation should be flexible to ensure responsive- important foundation on which to build. Project ness to community needs and rapidly changing targets and implementation periods may then conditions on the ground (IEG 2005). Overall, be planned and designed on the basis of such natural disaster response projects should seek to knowledge, rather than on “normal� capacities. achieve complex objectives using the simplest design that can deliver their objectives. Alternative networks, such as NGOs and United Nations agencies, can fill gaps, but their partici- Although many projects have effective design pation should be coupled with a plan to increase approaches, many still have had unrealis- government administrative capacity (IEG 2010a). tic objectives. Across some 60 disaster activi- Design should limit the number of implementing ties reviewed in the IEG Hazards of Nature agencies and the number of sectors involved, evaluation (2006), most required extensions of as well as reduce the conditions placed on the about a year and a half on three- to six-year lending. projects, and many of the extended projects still did not achieve their original targets. Many Response projects (or response elements conventional (nondisaster) projects have similar of larger multiphase projects) need to overruns; such “delays� may represent a realistic be developed and implemented rapidly, approach to incorporating longer-term activities using streamlined procedures wherever into emergency recovery loan (ERL) activities, possible. which were formerly restricted to three years for implementation. It is essential that project design The importance of speed for disaster response take into account the fact that the borrower on all activities has been emphasized in a note for the levels—local and national government, as well Haiti earthquake (IEG 2010c), which highlights as communities—will have a diminished capacity the value of the following disaster relief to function following a disaster. This realization elements: should be coupled with a clear analysis of the borrower’s institutional capabilities on all levels. • Streamlined decision making and procedures Such an analysis should be part of the design for contracting civil works to help avoid de- phase and feeds into the next lesson. lays. • Leveraging existing private sector capacity as Project and program deadlines should be critical for effective emergency response. based on a specific assessment of national • Working with International Finance Corpora- and local capacity in the post-disaster tion (IFC) clients near affected areas to gain situation, not on “normal� circumstances. speed and effectiveness. In such situations, matching grant schemes can be powerful in- In emergency situations, when a government’s struments for emergency response, but their capacity is adversely affected by the disaster, it effectiveness can be limited by the nature of is even more important not to set overambitious IFC’s pre-existing activities. 12 natural Disaster response: lessons from evaluations of the WorlD bank anD others • Balancing speed with careful assessment of disbursement regulations. IEG’s natural disaster demand and relevance in rapidly changing evaluation (IEG 2006) found that budget support post-disaster conditions. operations were not any quicker than ERLs for transferring resources. The importance of rapid action in promot- ing longer-term mitigation and risk reduction Although essential to effective implemen- measures (which can merge into a future Phase tation, the rapid development and process- 1 program) has also been demonstrated. An ing of disaster-related interventions present IEG lessons review (IEG 2005) emphasizes the ownership challenges. This is particularly importance of reaching agreement with the so in situations where several donors and government on mitigation measures within the partners are attempting to activate their first three months, because it gets much harder assistance at the earliest opportunity, such to get politicians to focus on disaster once the as in the international responses to the memory of the emergency recedes. Once these Haiti earthquake and the Asian tsunami. agreements are reached, they need to be locked into some form of public commitment, includ- The responses to many disasters have raised ing on financing mechanisms, to which people major concerns about the extent to which can refer to keep the government on track. local institutions and communities were bypassed in planning and implementing activi- However, speed in and of itself is not enough. ties. As a recent major example, the interna- One of the immediate post-disaster challenges tional response to the Haiti earthquake raised is to respond to urgent reconstruction needs, many questions concerning the effectiveness but in ways that improve on past practices of the assistance offered by the international and reduce the chance of a recurrence of community. For example, a structured synthe- problems in the future. Although initial quick sis evaluation (University of Haiti and Tulane action is important, experience suggests that University’s Disaster Resilience Leadership in an emergency situation, such as that which Academy 2011) reports that, although short-term prevailed after major floods in Pakistan, subproj- economic activities have been introduced and ect readiness should not divert the invest- are ongoing, there has been little reflection as to ment focus from a well-planned priority list whether the protracted emergency response has (IEG 2010b). Therefore, if the highest priority supported or has actually undermined resilience programs are not ready to implement, it is better in Haitian communities. Twelve months after not to launch lower-priority activities simply the devastating January 2010 earthquake, more because they are ready. Similarly, although than 1 million Haitians remained in camps, and damage assessments are quick, detailed, and substantive recovery had not yet begun. focused, they are not one-off efforts. They should be updated as the situation develops. Many of the evaluations reviewed in this analysis Quick initial action and a realistic schedule identify the lack of Haitian participation in are often major success factors in responses to decision-making processes as a major concern natural disasters. The actions of the first few and obstacle to building individual, household, days affect all future decisions. community, and national resilience. Moreover, resources and activities still focused on humani- Emergency projects require special attention tarian maintenance work, and the notion of to disbursement arrangements (IEG 2010a). “building back better,� had not yet penetrated Bottlenecks to cash flow should be minimized relief and recovery efforts. A key finding is the before project approval through provision of need for more engagement of Haitian local guidelines, sample bidding documents, techni- leaders, civil society, and—more importantly— cal assistance to first-time borrowers, training those directly affected in future project develop- in procurement procedures, and simple local ment to promote resilient recovery.1 13 evaluation brief 16 Similar issues were faced in connection with The importance of encouraging national the Asian tsunami response. An international ownership should be coupled with attention evaluation (Bennett and others 2006) finds to the concept of inclusion. The engagement that a consequence of the “swamping� of local of a very diverse range of national and interna- capacity by the large international presence tional stakeholders and activities has been in Aceh (Indonesia) and Sri Lanka was poor highlighted as a key feature of successful natural representation of local NGOs and community- disaster support programs. In Bank-supported based organizations, and little consultation with natural disaster interventions, the private sector them in coordination meetings. This may have has emerged as an important stakeholder and led to the erosion of local emergency capacities. potential partner, particularly after disaster has An IEG review of recent Bank experience (IEG struck. However, as with all partnerships, care 2010a) notes that rebuilding social structures is needs to be taken to match the intended role with a large challenge and one that is rarely done actual capacity on the ground. Reconstruction well. The characteristics of initial disaster initiatives, especially commercial ones, need to response make this even more difficult if the be carefully assessed against changes in effective response ignores local institutions and creates demand in the context of aid flows and rapidly dependence. Even in community development changing post-disaster conditions (IEG 2010a). interventions, where participatory approaches are inherent, implementation can still break For example, IFC facilities established to support down, especially when there are pressures to private sector companies in the reconstruction rebuild structures rapidly. phase of the Asian tsunami disaster were used only to a limited extent because their pricing Ideally, the engagement of local stakehold- was not attractive, given the abundant liquid- ers can be assured through their participation ity in the market and aid money pouring into in long-term disaster management activities. the affected countries. Local banks in Thailand An evaluation for the European Commission and Sri Lanka received cheap long-term funding (Aguaconsult Ltd. 2009) presents some key from their respective governments; the larger aspects of an approach founded on local companies had adequate insurance coverage ownership, as shown in Box 3. to repair/reconstruct their damaged properties; and most companies scaled down their new Natural disaster response projects need to investments, thereby reducing their need for include a broad range of national and local additional funds. stakeholders, particularly the poorest and the most vulnerable, as well as the private Leveraging existing private sector capacity sector. is critical for effective emergency response, Box 3: Some Factors Likely to Enhance Local Ownership of Disaster Response Projects • Knowledge and skills are more likely to be retained if the project • It is critical to get commitment from local partners and other includes activities that communities can put into practice. For participating stakeholders prior to commencing the project. example, communities can practice mobilization skills and This helps ensure everybody is clear on roles and responsibili- organization of events by implementing community workdays ties and avoids delays in short project time frames. or simulation exercises. • Quality baseline data, including participatory needs assess- • Coordination with government and local stakeholders is critical ments and a historical understanding of other similar inter- for legitimacy and sustainability. All local level initiatives must ventions, are a critical component of building relevant and link to the larger national disaster management system. sustainable response projects. Source: Aguaconsult Ltd. 2009. 14 natural Disaster response: lessons from evaluations of the WorlD bank anD others even where such capacity is quite weak. The population displacement and the prolonged private sector can play a key role in infrastruc- existence of temporary shelter settlements ture and logistics, local banking, and provision (Refugees International 2010). of physical capacity. Working with existing IFC clients near affected areas can bring major The importance of incorporating gender gains in speed and effectiveness. Such existing dimensions in disaster management has also partners do not need screening for reputational been emphasized (Bennett and others 2006). risks and delivery capacity. Trust and familiar- The dearth of gender-disaggregated data has ity allow the use of simple arrangements for impaired the effective targeting of vulnerable payments and reimbursements. Partners with groups and reinforced discriminatory practices. local presence have the knowledge necessary The overall Asian tsunami response lacked a to ensure that help reaches intended benefi- consistent, quantified, and coordinated gender ciaries (for example, a local bank in Sri Lanka analysis, an omission that has resulted in some was effective in directing livelihood restoration serious protection anomalies and the persistence grants to local fishermen who had lost their of male-dominated decision-making structures boats in the tsunami). that have largely gone unchallenged. In particu- lar, gender disaggregated data on which to base “Piggybacking� on existing activities is also seen targeted programs were largely missing in both as advantageous in the case of social funds, the relief and recovery phases. whose institutionalized operational procedures provide an immediate implementation platform Damage assessment often overlooks the that can reach stakeholders who may otherwise gender dimension of vulnerability (IEG 2010b). be overlooked. Having an existing public works Disasters may present opportunities to enhance program running under the social fund allowed gender equality, but there are gender-related for a fast response to crisis in a low-capacity social constraints in many countries that contrib- institutional setting in Tanzania (World Bank ute to slow progress. IEG has reported that it 2009c). is often the case that the impacts of disaster are uneven and that marginalized groups need The main socioeconomic category highlighted special attention (IEG 2010a). An awareness of for inclusion in disaster management activities is this issue should be built into damage assess- vulnerable people. For example, an international ment processes, so that they take into account NGO report indicates that, as the focus turns to the differential effects of disasters according to rebuilding large areas of Pakistan, donors and income, culture, gender, location, type of home, assistance providers should take the opportunity and land tenure. If this is not done, the particu- for reconstruction efforts to address the underly- lar needs of the poor may be ignored in the ing vulnerabilities of affected people (Refugees immediate post-disaster period and the vulner- International 2010). This can only occur by able may have to sell their productive assets, working closely with affected communities. The often including their land, to the better-off, as report notes that reconstruction funds often go happened following the tsunami in Indonesia first to landowners, large infrastructure projects, (IEG 2010a). industry, or developers. The need to provide homes, security, and livelihoods for the poor Partnership with NGOs can be important in and landless is often secondary, as is evident reaching the poor. However, it is important to from the experience of an earlier earthquake use adequate screening mechanisms to identify in the region. The International Federation of suitable NGOs. IEG’s review of community- Red Cross and Red Crescent Societies noted that driven development lessons from the Sahel after the 2005 earthquake in Kashmir, the failure (IEG 2003) finds that in Benin, NGOs with of authorities to provide housing assistance to poor qualifications handicapped project perfor- the landless was a major reason for long-term mance, and in both the social fund and the food 15 evaluation brief 16 security projects a large number of NGOs had to design should allow for a longer period of be suspended for unacceptable performance. implementation, initially focusing on rehabil- itation and later on disaster risk mitigation. Even though pressures to act quickly are The later phases should build on institutional strong after disastrous events, it is vital to and community resources mobilized during incorporate future disaster preparedness recovery and rehabilitation. into recovery and rehabilitation programs (designed during the response phase). This lesson shows that the different phases of disaster management are conceptual and As noted above, post-disaster situations provide organizing principles, rather than chronological unique opportunities (which conceptually fit sequences. In fact, it can be seen that there are within Phase 1 activities) to mitigate elements of considerable advantages in building on disaster disaster risk by preparing communities to better response programs to commence pre-disaster deal with them. Communities are particularly planning for any future event. This is because focused on such issues and are willing to act after disasters provide a short but intensive window of disasters. Furthermore, government resources attention for disaster risk reduction and mitigation are mobilized, creating a uniquely favorable opportunities, which fades away as other develop- environment for implementation of such ment priorities regain their higher profile. disaster preparedness components. However, the Sri Lanka Tsunami Emergency Recovery Faced with the competing demands of rehabili- Program missed this opportunity (World Bank tation/reconstruction versus long-term risk 2009g). Disaster preparedness elements were mitigation, the Bank has often focused on not built into the initiatives. Even though those reconstruction. This approach can delay or coastal areas where the mangrove forests and reduce the implementation of risk mitigation coral reefs remained intact were less impacted programs. by the tsunami than areas where prior environ- ment denudation had taken place, no measures ICRs of two Indian emergency recovery were taken to include disaster mitigation and projects (World Bank 2003, 2009b) find that preparedness initiatives, such as mangrove the integration of long-term goals (such as reforestation, in the program. These could have disaster management capacity building, drafting provided a unique opportunity to maximize the of seismic resistance planning standards, and “build back better� approach. ensuring quality control in dams reconstruction) into a disaster reconstruction project is likely to Following the torrential rains and flooding in increase its effectiveness, but requires a longer Algeria in 2001, the Bank supported the Urban (and more realistic) implementation period. Natural Hazard Vulnerability Reduction Project. As well as the large scale of such operations, This established and implemented a number nurturing and restoring the confidence of the of regulations and measures concerning urban affected population are time-consuming activi- infrastructure, which were expected to reduce ties, although they play an important role in the effects of any future flooding, particularly in ensuring ultimate cost effectiveness. Algiers (World Bank 2007a). Although there is no evaluation of the results of these measures, In the reconstruction in Gujarat, the project they illustrate that the Bank is now active in required revision of planning and building codes disaster risk reduction, drawing on its earlier to ensure that structures would be earthquake experience in the field of response. resistant. Although this caused initial delays in housing reconstruction in urban areas, the To incorporate future disaster preparedness process speeded up over time as masons, into recovery and rehabilitation programs engineers, and technical officers were trained (which fall into the response phase), program and became familiar with the new requirements. 16 natural Disaster response: lessons from evaluations of the WorlD bank anD others This suggests that, even within the project time of proper handling of disaster information. frame, the effectiveness of disaster risk reduction had increased. Advances in information and communications technologies represent opportunities to create Information, communication, and data new solutions, including, for example, systems management systems have a vital role to that allow remote units to enter data directly in play in disaster-related projects. However, the system using cellular or satellite networks. they have often been found wanting. There is a potential to achieve exponential gains in the efficiency of disaster response operations. Effective, consistent, and coordinated communi- However, it is sometimes the case that simpler cation with recipient populations at all stages technologies are more flexible in an emergency of the response—and with a concerted effort situation, such as the use of portable flash drives to include women in the dialogue—needs to to share spreadsheets in Mozambique (IEG, be made a priority (Bennett and others 2006). undated). This entails dedicated staff resources and tools, with efforts made to reach a communications In the case of the Asian tsunami, the Tsunami protocol with the host government. A common Evaluation Coalition found that cell phones and strategy needs to be developed, including satellite imagery emerged as important instru- the use of public meetings, broadcast media, ments of communication and coordination in newsletters, and posters. The creation and use the immediate stages of the emergency (Telford of a common beneficiary database, financed and Cosgrave 2006). Much of this technology and endorsed by a central government body, was in the hands of the national private sector, is another early priority in the emergency emphasizing the importance of greater efforts phase. Required leadership and coordination to develop partnerships between local and skills include the basics of how to maximize international stakeholders, public and private, the output of meetings. These skills should be to improve the quality of available information promoted by all agencies and should form part and the speed of its delivery. of the induction training for operational staff, along with standard operating procedures. The Office for the Coordination of Human Affairs and World Bank report (OCHA and The necessity to encompass the most appropri- World Bank 2008) also highlights the increasing ate technology within a coherent institutional global recognition of the need to take the step structure has been emphasized in an assess- from ad hoc disaster responses to the system- ment jointly produced by the United Nations atic ex ante development of disaster manage- Office for the Coordination of Humanitarian ment infrastructure by vulnerable countries Affairs and the World Bank (OCHA and World or provinces and districts at risk. Despite this Bank 2008). An important lesson that emerges recognition, few well-functioning systems for from the report’s case studies is that an effective information sharing during disaster response disaster information management system periods have been developed. requires a good technological platform, but also much more. Software programs for storing, Projects responding to natural disasters sharing, and manipulating data for disasters are have proved difficult to finance effectively. being developed at a steady pace, often in the Both ERLs and reallocation of existing aftermath of disasters. The real difficulty lies in resources have encountered problems in anchoring these technological approaches in providing funds quickly enough or for long an appropriate institutional context, where they enough to achieve all their objectives. are supported by relevant and effective operat- ing procedures; agreed terminology and data The World Bank’s main instrument to support labeling; and a shared awareness of the benefits activities related to natural disasters is the ERL, 17 evaluation brief 16 which has a maximum duration of three years. ally proves to adversely affect the programs This is sometimes seen to impose restrictions on from which the money has been removed.2 the planning of support, because reconstruc- Against these negative assessments, realloca- tion can take substantially longer than this tion of funds fills an important niche in the period. Reallocating resources from existing Bank’s ability to respond rapidly,3 particularly in projects, another approach to emergencies, large emergency situations, where the original has been found to affect the ability to attain purpose of the loans may no longer be viable in long-term development goals and to be less view of the disaster (IEG 2010b). effective than specific reconstruction lending (IEG 2005). Notes 1. It can, however, be suggested that the govern- Although shifting resources from existing mental and institutional situation in Haiti was an programs to rehabilitation and reconstruction unusually difficult one in which to launch participa- efforts with very high rates of return can be tory approaches. justified, experience has shown that new financ- 2. IDB’s 2004 evaluation, Evaluation of Inter- ing, well designed and managed by special American Development Bank’s Operational Policy disaster units that are authorized to respond on Natural and Unexpected Disasters, found major quickly, tends to be used more effectively (IEG problems with loan reformulation for disaster-related 2010c). Restructuring old projects is often politi- projects and recommended measures to ensure that cally easier than new lending and allows the disaster management should become a high priority Bank to support government entities that are issue in national development programming. already accustomed to working with it, but 3. IDA countries are not eligible for CAT-DDO status, delivery by staff committed to the goals just which gives access to the most rapid fund transfer abandoned is often not effective and eventu- system. 18 5. Lessons for the Post-Disaster Phase (Phase 3) T he post-disaster phase includes activities in recovery, rehabilitation, and reconstruction, often building on preparation activities undertaken during the previous response phase. It also affords an opportunity to develop disaster risk reduction measures, which can be fully implemented during a subsequent pre-disaster phase (moving into Phase 1 of the next cycle). Post-disaster recovery may best be pursued success was that the Authority was formed with by strengthening existing institutions or by a time-bound yet clear mandate and was staffed creating time-bound specialist bodies with through existing government resources, with the sole mandate of completing short- to additional key experts hired through donor medium-term reconstruction activities. assistance. It was, therefore, seen as a govern- ment-led, but donor-assisted, entity having the It is difficult to create effective new institutions in full support of other implementing government the aftermath of a disaster, unless their mandate line agencies. and duration are defined very specifically. For the Sri Lanka Tsunami Emergency Recovery If existing institutions are seen as the best option, Program (World Bank 2009g), the creation of weaknesses, gaps, and capacity constraints in a new institutional set-up in the aftermath of the overall functioning of government structures the disaster to implement reconstruction efforts need to be strategically addressed in advance of provided major challenges. These concerned major recovery and reconstruction efforts.1 This the ability of the new body to take on the process should include attention to estimating reconstruction task in such a short period of and agreeing on incremental staffing increases time, as well its sustainability. The Reconstruc- before the program is implemented on a large tion and Development Agency received a scale. This should be based on estimates of large amount of funding from donors but incremental increases in workloads on program could not build capacity to its optimum level staff at national, regional, and local levels. and was ultimately prematurely closed down by the government and its functions given to A community-driven approach to recovery the Ministry of National Building and Estate and reconstruction can significantly help Infrastructure Development. build up local capacities for future project identification, planning, and implementa- In a similar situation in Pakistan (cited as a tion (including procurement and financial comparative example in World Bank 2009g), the management), as well as in ex post opera- Earthquake Reconstruction and Rehabilitation tion and maintenance. Authority, which was created in the aftermath of the 2005 earthquake, performed well in Such an approach need not be expensive. developing capacity and effectively respond- For example, the Samoa Cyclone Emergency ing to the disaster. A factor contributing to this Recovery Program (World Bank 2009f) found 19 evaluation brief 16 that the size of community grants used for lack of understanding of the program and its identifying, planning, and undertaking both objectives, which translated into lack of commit- structural and nonstructural solutions need ment in implementation support. This aspect of not be large. What proved more important in the Sri Lanka program shows that information this program was the participatory process by dissemination and social mobilization need to which appropriate solutions were identified, be treated as priority components to ensure so that there would be continuing community that all potential participants actually have the commitment to provide additional resources opportunity to benefit from Bank support. and to ensure that the assets developed were maintained and well used after the “project� In recognition of the importance and difficulty closed. Central government ministries played of encouraging strong community engage- an important role in this regard, either directly ment, the Bank has provided detailed advice through their own structures or through NGOs (World Bank 2009b) on appropriate methods, and community-based systems (or both). This particularly through the use of social funds and included building interest in and “marketing� the community-driven development approaches. availability and purpose of grants; ensuring that the grants were easily accessible by communi- In cases of major reconstruction of ties for eligible activities; providing engineer- communities following a natural disaster, ing/environmental and other technical inputs an approach that gives maximum ownership to communities at key points; and encouraging and control to homeowners themselves has communities to take ownership of the eventual major advantages over the contractor-led assets. system (Box 4). In contrast to the Samoa project, the Sri Lanka IEG has reported that allowing homeowners to Tsunami Emergency Recovery Program was manage the reconstruction of their own homes hampered by inadequate social mobilization. (rather than engaging contractors) worked well Although under its housing program there was in two India projects (World Bank 2003, 2009b), provision for formation of village rehabilitation as well as in Turkey (World Bank 2007d). In committees, there is little evidence to suggest India, where people were given funds to that this was widely practiced or that, where repair their own housing units, most families formed, these committees significantly contrib- actually economized enough to self-build uted to mobilizing communities. NGOs involved completely new homes. In the Sri Lanka in the housing program also did not have Tsunami Emergency Recovery Program, the social mobilization as a primary focus area, but approach adopted was driven by homeowners concentrated on contractor-driven reconstruc- themselves, supported with necessary technical tion under a donor-driven housing program. assistance. This proved a success and allowed affected communities to develop ownership of A continuous information dissemination the program, which yielded strong results. Many campaign was envisaged under the overall beneficiaries stated a preference for the system strategic communications vision of the Bank’s adopted by the Bank, rather than the contractor housing component, but this campaign did not construction method used in a parallel donor- materialize. There were a few housing informa- driven program. Homeowners under the Bank tion products, such as frequently asked questions approach had the incentive of constructing and posters, developed on an as-needed basis, houses of their own choice and specifications but there was no overall effective dissemination within safety guidelines defined by the program. plan. They also received useful training in construc- tion and supervision, and their participation in Among other effects, the limited strategic the reconstruction of their houses gave a strong interventions in communications resulted in a sense of ownership and empowerment.2 20 natural Disaster response: lessons from evaluations of the WorlD bank anD others Box 4: Key Advantages of the Homeowner-Driven Approach to Reconstruction • Ownership by affected communities of housing solutions • Faster pace of reconstruction (interest of beneficiary) • Greater room for inclusion of beneficiary choices (especially • No artificial inflation of material prices because of reduction concerning architectural aspects) in pressures on the materials supply chain through stag- • Promotion of safety culture and awareness gered construction (especially in post-disaster reconstruc- • Skill development of affected communities tion situations). Source: World Bank 2009g. Although programs that support homeown- The program could have been improved if ers in reconstructing their own houses cofinancing opportunities had been available to have been found broadly effective, they mitigate increases in construction costs and/or if need to incorporate specific systems to partnerships had been formed with humanitar- identify and support vulnerable categories ian organizations to support vulnerable families of the population to ensure that this group and to monitor and support “nonstarters.� is reached. By their nature, emergency recovery and The Sri Lanka Tsunami Emergency Recovery reconstruction projects and programs Program aimed to provide assistance to eligible are likely to be prepared under consider- homeowners to repair or reconstruct their able time pressure, during the medium- homes. Although the targeting achieved through term response phase. This haste should this process was broadly effective (World Bank not lead to a reduction in systems to 2009g), it became apparent that there was a ensure accountability and transparency of need to provide special assistance to vulnera- programs, particularly because these activi- ble groups, such as female-headed households ties may disburse large sums in a short time and the elderly, so that they could reconstruct period. their houses in a timely manner. Without this attention, these groups appeared to be disadvan- In its responses to the Asian tsunami, the taged. However, the project did not have any Bank promoted a range of measures to special arrangements to cater for this particular increase transparency. In view of the risk of segment of the affected population. funds misdirection, participatory community monitoring, such as that used in Indonesia, The (India) Gujarat Emergency Earthquake may be valuable. In the Indonesia Kecamatan Reconstruction Project also found that its development program and tsunami emergency homeowner-driven approach was popular projects, action was taken to minimize corrup- and effective in assisting reconstruction. The tion by posting planned works and costs on high levels of satisfaction among beneficiaries village notice boards; having village committees pointed to the perceived and actual benefits of audit neighboring villages; having communi- the homeowner-driven program. However, its ties physically sign off on contractor quantities assumption that all homeowners could build delivered; and putting in place a strong group within a particular target period excluded to deal with (anonymous) complaints. beneficiaries who were unable to meet the program’s deadlines. These beneficiaries then The Philippines Bicol Power Restoration Project faced increased construction costs (because of provides another good example of how borrow- inflation), which had a further negative effect ers and the Bank can efficiently and effectively on their completion rates. They thus became work together to ensure acceptable account- doubly disadvantaged. ability procedures during the response to an 21 evaluation brief 16 emergency situation (World Bank 2009d). The project was having housing fund disbursement project was based on a clear assessment of the linked to construction progress (for individual capacity of the main participating institution, housing); this helped ensure effective use of TransCo (a government-owned power transmis- funds. Without this link, there was a risk that sion company), to satisfy Bank accountability beneficiaries might use funds for purposes other requirements and on the establishment of agreed than their intended use or at least be perceived processes, which resulted in fast and generally to do so. problem-free implementation. Transparency and accountability are not only Overall, four aspects of the approach contributed important for each institution offering assistance. to the success of the project: (i) effective collab- The relationship between the programs of differ- oration with a receptive institution; (ii) flexibility ent international stakeholders is also extremely and responsiveness, but applied within a careful significant. If communities and households with assessment of agency capabilities and clear similar needs are treated differently by various application of financial and governance require- programs, major problems can arise. In this ments; (iii) effective safeguards management; respect, accountability to the affected popula- and (iv) complementarities of the assistance tion is a cornerstone of good coordination with sector reform processes. Accountabil- practice (Bennett and others 2006; Telford and ity mechanisms were addressed in advance Cosgrave 2006). In the case of the Asian tsunami, through careful review of TransCo’s procure- effective joint-agency communication with ment processes and determination of flexible the client population, including a complaints arrangements that met Bank standards. After procedure and regular updates on the recovery early assessment, financial control procedures process, was not an early priority. Communica- were put in place through a well-staffed internal tion and consultation between the international audit team that would bring the project within its community and affected people was sporadic ambit and collaboration between the Commis- and uncoordinated, leading to widespread sion on Audit and TransCo to address anomalies misinformation and resultant frustration among in accounts that would be resolved prior to stakeholders. Bank appraisal of the next potential project with TransCo. In any disaster-related activity, it is important to ensure from the outset that The India Gujarat Emergency Earthquake appropriate systems are put in place to Reconstruction Project is another project that deal with complaints and grievances. shows that transparency, equity, and accountabil- ity can be achieved, even in difficult emergency Even if accountability and transparency circumstances, if appropriate procedures are mechanisms are effective, there remain possibili- designed and incorporated as integral parts of ties that some eligible people, particularly the overall intervention program (World Bank among vulnerable groups, may find it difficult 2009b). The project supported the repair and to access intended resources and that grievances reconstruction of over 1 million houses. This may be raised against the intervention. The was achieved with minimal grievances and Bank’s Sri Lanka tsunami program included a allegations of corruption, reflecting success- robust grievance redressal mechanism (GRM), ful incorporation of transparency and people’s under which village-, district-, and divisional- voice. A particularly important element was level Grievance Redressal Committees were the approach adopted for damage assessment. established to deal with land issues and other Although a “technical� exercise, this benefited housing related complaints (World Bank 2009g). from substantial community participation, To deal with grievances related to housing which helped ensure transparency and social damage, special district teams with the required acceptance. Another important aspect of the technical skills were formed. Part of their work 22 natural Disaster response: lessons from evaluations of the WorlD bank anD others was to reassess the value of houses, whose needs, which changed frequently as the disaster owners objected to the original assessment situation unfolded. The ICR concludes that the made by the damage assessment teams. The flexible approach used by the project worked district and damage assessment teams comprised well and would be beneficial to the design of different sets of officials, in order to avoid any other emergency operations. conflict of interest. Despite these efforts, during implementation the GRM did not function to its The Grenada Caribbean States’ Emergency full potential, mainly because of poor documen- Recovery and Disaster Management Program tation of grievances and the manner in which the (World Bank 2006a) showed the value of regional housing units were wound up at the end flexible emergency procurement procedures of the program. However, the GRM did serve the in the context of urgent recovery needs. The purpose of increasing accountability and provid- length of time before switching from flexible ing the beneficiaries with a framework through to regular procurement methods should depend which to resolve their complaints. Some require- on the magnitude of the disaster, the damages ments and benefits of such systems are shown incurred, and the pace of implementation on in Box 5. the ground. The Grenada project successfully demonstrated this principle, as procurement Flexibility, especially with regard to processes reverted to standard procedures two procurement procedures and project years after the hurricane hit, adapting to the revision as post-disaster events unfold, is a situation in the country. particularly important attribute for natural disaster-related projects. Because of the short preparation time and the nature of emergency operations, ERLs are riskier Flexibility of project design in disaster-related than standard Bank operations. This is particu- operations facilitates implementation and larly true concerning expediency and controls provides a greater opportunity for achieving over procurement processes, and it is important project objectives. In the Sri Lanka Tsunami to consider measures that mitigate risk. One such Emergency Recovery Program, the project design measure is having a procurement consultant on was kept flexible (World Bank 2009g). A number staff closely following the project throughout its of contingent components were included, with lifetime. This calls for a relatively large supervi- the idea that the scope of work and the areas sion budget, a good percentage of which should of intervention would be adjusted as project be available to pay for a procurement special- implementation proceeded, based on identifica- ist. This approach was adopted in the Iran Bam tion and prioritization of emerging needs. It was Earthquake Reconstruction Project, and the ICR prepared on a fast-track basis under emergency reports that this contributed substantially to risk procedures, and it was not possible to cater for all reduction (World Bank 2010a). Box 5: Requirements and Benefits of Effective Grievance Redressal Mechanisms Requirements Benefits • Structured documentation, compilation of grievance data • Speedier resolution of grievances, leading to faster disburse- • Grievance tracking and redressal (response-time) monitoring ments and program completion systems • Progressive incorporation of beneficiary and community feedback • Central oversight and control of grievance redressal functions, in the project implementation framework and arrangements and decentralized grievance resolution arrangements with • Reduced risks to reputation emanating from lack of beneficiary local participation. satisfaction. Source: World Bank 2009g. 23 evaluation brief 16 Another important point is that streamlined It is important that natural disaster-related procurement processes in emergency situations project deadlines should be placed within a are needed on both the country and the Bank clear “results framework� to promote effective sides to ensure faster delivery times. Although the project implementation (IEG 2011a). This Bank may adopt fast-track procedures internally, framework would include the scope and borrowing countries often do not have correspond- specificity of objectives, congruence between ing emergency procurement processes. project interventions and objectives, links between objectives and measurable indicators, Other flexible procurement measures, which and effective M&E. have contributed to results of natural disaster projects, include the following: In the case of the Sri Lanka Tsunami Emergency Recovery Program, reporting and monitoring • Hiring key borrower project staff well in ad- were carried out at various levels, but they were vance of project start-up and providing them not part of an integrated reporting, monitoring, with training in Bank procedures, to enable and evaluation system. Although regular report- the project to move quickly on approval ing was undertaken at the regional level, the (World Bank 2010a). upward flow of data to bodies where meaning- • Using “slice and package� approaches for ful evaluation was supposed to take place, construction contracts. Under these, a large promoting the informed downward flow of homogeneous project is sliced into smaller strategic direction, was not fully realized. Such similar contracts that are bid simultaneously a system could have assisted the government in to attract the interest of both small and large even better implementation of the project. The firms. Firms may offer bids on individual ICR derives the lesson that an effective report- contracts (slices) or on a group of similar ing M&E system is essential to enable informed contracts (packages), and award is made to decision making at both policy and operational the combination of bids offering the lowest levels. Although this lesson appears to be realis- cost to the client (World Bank 2009f). tic, it is derived from the unsatisfactory results • In countries where agencies have no incen- achieved without such a system, rather than tive to submit invoices or to identify eligible from an evaluation of the operation and contri- investments, using small subprojects with bution of such a system as practiced. simplified financial management processes, especially those related to the validation of In the Zambia Emergency Drought Recovery disbursements and expenditures and to the Project (IEG 2007b), efforts were made to translation of these expenditures into Bank address the limitations of the initial project disbursements (World Bank 2005b). indicators during the first implementation review mission. The project implementation unit was Even more than “conventional� develop- subsequently tasked with identifying additional ment projects, interventions dealing with project performance indicators but ultimately natural disasters need strong monitoring failed to complete the task. and evaluation (M&E) systems with a clear results focus. Monitoring that is independent The Iran Bam Earthquake Emergency Recon- and credible is essential to give beneficia- struction Project (World Bank 2010a) provided ries confidence in the government’s support a positive example in this area. Even though and to demonstrate to donors that funds are emergency operations do not currently require being well spent. These measures are partic- outcome indicators, the project invested in an ularly important in such situations, because M&E consultant and a social auditor, which was the rapid expenditure of large sums may lead uncommon in similar projects. The ICR notes to specific concerns about how their effects that this reduced the risks to reconstruction can be systematically tracked. activities, which were felt to be strong because 24 natural Disaster response: lessons from evaluations of the WorlD bank anD others Box 6: Guiding Principles for Monitoring and Evaluation in Natural Disaster Interventions • Define and agree with stakeholders what will be monitored and provides agencies with guidance on the indicators it wishes to evaluated early in project development. be monitored at the project level. The indicators to be monitored • A mix of qualitative and quantitative approaches is likely to should be based on the reconstruction policy. be the most useful for M&E in a post-disaster situation. Par- • Good M&E principles are not different in a post-disaster ticipatory performance monitoring and small-scale household situation, but to apply them may require more flexibility and surveys are two especially useful qualitative tools. imagination. • Damage assessment data are a critical source of baseline in- • If government is not prepared to aggregate data collection from formation for evaluation, another reason to promote the sharing multiple agencies to monitor reconstruction, agencies in one of this information among agencies. sector or region should consider coordinating the monitoring • Government can simplify the task of tracking reconstruction if it among themselves. Source: World Bank 2010c, chapter 18. of the limited experience of government both in Notes such major programs and in working with the 1. These issues were recently discussed at the World Bank. A robust M&E system was designed and Reconstruction Conference (http://www.wrc-2011.org). implemented and contributed to the fulfillment 2. The homeowner-driven approach may be less of reporting requirements to the government effective in urban than in rural housing situations. and the Bank. This experience shows there are Recent discussion of these issues took place at the measures that can be taken to develop outcome World Reconstruction Conference 2011. indicators and to ensure that relevant data are 3. This guidance appears to have been derived from collected to verify progress. a broad review of international best practice in the disaster management field, including the World The Bank has provided specific guidance for M&E Bank’s experience, rather than from specific evalua- approaches in disaster related projects (Box 6).3 tion sources. 25 6. A Lesson Concerning Slow-Onset Disasters S low-onset disasters, such as droughts (which often become recurring events), are a subcategory of natural disasters that pose specific chal- lenges. There is a need to develop and implement strategies that deal with development and humanitarian issues at the same time. This requires collaboration between different types of agencies, which have traditionally found it difficult to work together. The European Commission conducted a detailed substantial “mission creep� of humanitarian review of responses to the Sahelian droughts agencies into recovery and reconstruction areas (ECHO 2007). It found that the prevailing model normally regarded as the territory of develop- in the Sahel of moving from development work mental agencies. The case of the Sahel droughts to emergency methods and back to develop- showed that the divide, which has been seen ment work each time a crisis arrives has not as a barrier to good development work for worked and is no longer appropriate to the years, remains unresolved. Rather, it has been situation of vulnerable people in the region. In reinforced by separate budget and administra- their haste to respond to disasters, humanitarian tive systems and, more importantly, by different agencies have often bypassed local government institutional cultures and strong personal identi- structures and provided relief directly to local fication with one side or the other. According to people, with little regard for local or national- the European Commission study (ECHO 2007), level representatives, thereby undermining the the realization that the two approaches need to role that these institutions play outside emergen- be made to work together is not yet widespread cies. Bypassing locally accountable institutions and is still principally found among personnel in favor of a more rapid impact has also allowed in some NGOs that already have operations in more powerful local stakeholders to “hijack� both fields. the process for their own benefit, which further reduces the confidence of local communities in The European Commission study proposes that the capacity of the structures meant to represent the adoption of an approach based on a sustain- them. Although investing and strengthening able livelihoods framework1 may offer opportu- existing local institutions would have taken nities for common work between humanitarian longer, it could also have reduced vulnerability and development institutions and between these in the longer term. departments within institutions. Such an approach “provides an analytical framework that promotes The traditional role of humanitarian assistance in systematic analysis of the underlying processes saving lives and helping populations get back to and causes of poverty. It is not the only such where they were before disaster struck proved framework, but its advantages are that it focuses less clear in the slow-onset disasters of the Sahel attention on people’s own definitions of poverty region, where it was also seen as necessary to and it takes into account a wide range of factors develop systems to reduce the exposure to risk that cause or contribute to poverty� (DFID-ODI of vulnerable communities. There was therefore 1999). An underlying lesson is that the basis for 27 evaluation brief 16 any planning in the Sahel is that drought will For example, in drought-prone areas, govern- happen at some stage of any initiative—during ments and the Bank have to be alert to the first planning or implementation or after completion. signs that farmers are selling animal herds and It follows that projects and programs should aim consuming seeds and must quickly intervene to to reduce both vulnerability and the impact of prevent the drought triggering a full-fledged food drought, instead of focusing on development crisis. Relief efforts for recurrent disasters (such objectives that fail to take account of the near as droughts) need to be built on data systems inevitability of future droughts. that are both accurate and current. Such systems may need technical assistance for several years, The World Bank has assisted several African before they are sustainable. Effective approaches countries to establish drought recovery systems. and funding options must therefore address both Assessments of these projects raise the same and short-term response and long-term recovery challenges, as noted by the European Commis- measures. sion review (ECHO 2007). For example, in the case of the Ethiopia Emergency Drought Assessments of the Zambia Emergency Drought Recovery Project further develop this perspec- Recovery Project, Bank assistance to the national system of drought response provided substan- tive (IEG 2007b; World Bank 2006b). They tial support to public works programs. These suggest that short-term and long-term objectives public works essentially performed two roles: should not be combined in one lending instru- relief, through cash transfers associated with ment. As in the other drought projects, activi- employment generated, and recovery, through ties fell between two types of interventions: the build-up of the (community) asset base (IEG crisis mitigation and long-term development. 2011a; World Bank 2007b). The ICR concludes The ICR indicates that this project would have that emergency contexts sharpen the trade-off been more successful if it had been more clearly between relief and recovery and that the need focused on crisis mitigation activities. It proposes to provide urgent relief may compromise the that, when engaging in emergency operations, technical quality of recovery activities or of the Bank should focus on its comparative arrangements for long-term “ownership� and advantage, which lies in such financial areas as sustainability. budget and balance of payments support. Such support would be designed to build the govern- Assessments of the Malawi Emergency Drought ment’s own capacity to respond to emergencies Recovery Project draw a similar conclusion (IEG by building and setting aside sufficient financial 2007a; World Bank 2005a). There is a conflict resources specifically for this purpose. between the ERL requirement for quick imple- mentation and the incorporation of medium- or Note long-term disaster management objectives. Food 1. For a brief overview of sustainable livelihoods crises can be prevented by early interventions. approaches, see Clark and Carney (2008). 28 7. Applying the Lessons S ince the publication of IEG’s 2006 evaluation on the subject, lessons have continued to emerge concerning natural disaster response activi- ties. These lessons have here been organized according to the phase of the disaster management cycle in which they first occur. One overriding lesson concerns the advantages Some of the lessons derived from project and of reducing risks and increasing preparedness program experience in the area of natural within the national development strategies of disaster response would apply to projects in any countries that are prone to disasters. Measures field, but are even more important in the case taken in pursuit of these objectives are believed of disaster projects, because of the situation of to deliver substantial gains in the effective- social, economic, institutional, and governmen- ness of disaster response, when events occur. tal disruption in which they operate. However, formal evaluations of the impacts delivered by risk reduction and preparedness There is also a distinctive set of lessons concern- are so far largely absent. ing the processes through which disaster response can be prepared and planned for in It is also clear that every natural disaster has unique advance. These relate to the professional area of characteristics, so that the lessons presented here disaster risk reduction within the overall disaster need to be carefully assessed for their applicabil- management cycle and to the need to raise its ity in each case. They provide a set of guidelines, profile both in affected countries and in interna- which can be examined for their relevance to tional organizations. Risk reduction should have specific disaster situations; this may help strengthen a central role in any disaster-prone country’s interventions and make them more effective and overall sustainable development strategy and efficient for affected countries, as well as for the should therefore form an integral part of donors’ Bank’s assistance programs. programs and country strategies. 29 Bibliography Aguaconsult Limited. 2009. “Evaluation of Disaster Operational Policy on Natural and Unexpected Preparedness of European Commission Humani- Disasters. Washington, DC: IDB. tarian Aid Department (ECHO) (DIPECHO) IEG (Independent Evaluation Group). 2003. Action Plans for the Caribbean.� http://www. “Community-Driven Development: Lessons from aguaconsult.co.uk. the Sahel—An Analytical Review.� IEG Working Amin, Samia, and Markus Goldstein (eds.). 2008. Data Paper, World Bank, Washington, DC. Against Natural Disasters: Establishing Effective ———. 2005. “Lessons from Natural Disasters Systems for Relief, Recovery, and Reconstruction. and Emergency Reconstruction.� World Bank, Washington, DC: World Bank. Washington, DC. http://www.oecd.org/ Bennett, Jon, William Bertrand, Clare Harkin, Stanley dataoecd/23/28/35281124.pdf. Samarasinghe, Hermantha Wickramatillake. ———. 2006. Hazards of Nature, Risks to Develop- 2006. “Coordination of International Humanitar- ment. Washington, DC: World Bank. ian Assistance in Tsunami-Affected Countries.� ———. 2007a. “Malawi Emergency Drought Recovery Tsunami Evaluation Coalition, London. http:// Project.� Project Performance Assessment Report www.alnap.org/resource/3530.aspx. No. 39849, World Bank, Washington, DC. Clark, J., and D. Carney. 2008. “A Review of DFID’s ———. 2007b. “Zambia Emergency Drought Recovery Experience with Sustainable Livelihoods.� Depart- Project.� Project Performance Assessment Report ment for International Development, London. No. 39847, World Bank, Washington, DC. http://www.eldis.org/assets/Docs/41798.html. ———. 2010a. “Responding to Floods in West Africa: DFID-ODI. 1999. “Key Sheets for Sustainable Liveli- Lessons from Evaluation.� IEG Note, World Bank, hoods.� Department for International Develop- Washington, DC. http://siteresources.worldbank. ment, Natural Resources Policy and Advisory org/EXTDIRGEN/Resources/flood_note.pdf. Department, London. ———. 2010b. “Response to Pakistan’s Floods: DEC (Disasters Emergency Committee). 2010. “A Evaluative Lessons and Opportunity.� IEG Note, Review of Disaster Risk Reduction Work by World Bank, Washington, DC. http://sitere- DEC Member Agencies in Response to the 2004 sources.worldbank.org/EXTDIRGEN/Resources/ Tsunami.� London, United Kingdom. http:// ieg_pakistan_note.pdf. www.alnap.org/pool/files/dec-drr-review-feb- ———. 2010c. “World Bank Group Response to the 2010.pdf. Haiti Earthquake: Evaluative Lessons.� Evaluation ECHO (European Commission, Humanitarian Aid Brief 10. Washington, DC. Department). 2007. “Drought and Vulnerability in ———. 2011a. “Ethiopia Emergency Drought Recovery the Sahel Region of West Africa: Concept Paper— Project.� Project Performance Assessment Report Final.� European Commission, Brussels. No. 62723, World Bank, Washington, DC. Global Alliance for Public Relations and Communi- ———. 2011b. “Ethiopia Productive Safety Nets cation Management. 2010. “Disaster Communica- Project.� Project Performance Assessment Report tion: Lessons from the Asian Tsunami Tragedy.� No. 62549, World Bank, Washington, DC. http://www.alnap.org/pool/files/GA_Disaster_ ———. Undated. “Note on Disasters in Mozambique.� Communciations_Lessons_200506.pdf. IEG Note. http://www.ieg.worldbankgroup.org. Inomata, Tadanori. 2006. “Towards a United Nations Norad (Norwegian Agency for Development Cooper- Humanitarian Assistance Program for Disaster ation). 2007. “Humanitarian Response to Natural Response and Reduction: Lessons Learned from Disasters: A Synthesis of Evaluation Findings.� the Indian Ocean Tsunami Disaster.� Joint Inspec- Oslo, Norway. tion Unit, United Nations, Geneva. ———. 2010. “Norwegian Humanitarian Response to IDB (Inter-American Development Bank). 2004. Natural Disasters: Case of Haiti Earthquake.� Final Evaluation of Inter-American Development Bank’s Report. Oslo, Norway. 31 evaluation brief 16 OCHA (United Nations Office for the Coordination Completion and Results Report No. 35429, World of Humanitarian Affairs) and World Bank. 2008. Bank, Washington, DC. “Uganda Floods: Lessons Learnt Workshop— ———. 2006b. “Zambia Emergency Drought Recovery Final Report.� Soroti, Uganda, January. http:// Project.� Implementation Completion and Results reliefweb.int/node/257013. Report No. 3492, World Bank, Washington, DC. ProVention/International for Institute Applied Systems ———. 2007a. “Algeria Urban Natural Hazard Analysis (IIASA). 2006. “Disaster Insurance for the Vulnerability Reduction Project.� Implementation Poor? A Review of Micro Insurance for Natural Completion and Results Report No. ICR000033, Disaster Risks in Developing Countries.� Geneva, World Bank, Washington, DC. Provention. ———. 2007b. “Ethiopia Emergency Drought Refugees International. 2010. “Confronting Climate Recovery Project.� Implementation Comple- Displacement: Learning from Pakistan Flood.� tion and Results Report No. 34926, World Bank, Washington, DC, Refugees International. http:// Washington, DC. www.alnap.org/resource/6077.aspx. ———. 2007c. “Ethiopia Productive Safety Nets Steering Committee for Humanitarian Response. Project.� Implementation Completion and Results 2010. “Peer Review on Accountability to Disaster- Report No. ICR000008, World Bank, Washington, Affected Populations: An Overview of Lessons DC. Learned.� Geneva. ———. 2007d. “Turkey Marmara Earthquake Telford, John, and John Cosgrave. 2006. “Joint Evalua- Emergency Reconstruction Project.� Implemen- tion of the International Response to the Indian tation Completion and Results Report No. Ocean Tsunami: Synthesis Report.� Tsunami ICR0000131, World Bank, Washington, DC. Evaluation Coalition, London. http://www.alnap. ———. 2009a. Building Resilient Communities. Risk org/pool/files/889.pdf. Management and Response to Natural Disasters University of Haiti and Tulane University’s Disaster through Social Funds and Community-Driven Resilience Leadership Academy. 2011. Haiti Development Operations. Washington, DC: World Humanitarian Aid Evaluation, Structured Analysis Bank. Summary Report. New Orleans/Port-au-Prince. ———. 2009b. “India Gujarat Emergency Earthquake http://www.alnap.org/pool/files/1228.pdf. Reconstruction Project.� Implementation Comple- UN ISDR (United Nations International Strategy for tion and Results Report No. ICR0000638, World Disaster Risk Reduction). 2002. “Living with Risk: Bank, Washington, DC. A Global Review of Disaster Reduction Initiatives. ———. 2009c. “Natural Disaster and Crises Response International Strategy for Disaster Risk Reduction.� by the Tanzania Social Action Fund.� In Building Geneva: United Nations. Resilient Communities: Risk Management and ———. 2005. “Hyogo Framework for Action Response to Natural Disasters through Social 2005–2015: ISDR International Strategy for Disaster Funds and Community Driven Development Reduction International Strategy for Disaster Operations. Washington, DC: World Bank. Reduction: Building the Resilience of Nations and ———. 2009d. “Philippines Bicol Power Restoration Communities to Disasters.� World Conference on Project.� Implementation Completion and Results Disaster Reduction. Kobe, Hyogo, Japan. Report No. ICR00001003, World Bank, Washing- World Bank. 2003. “India Andhra Pradesh Hazard ton, DC. Mitigation and Emergency Cyclone Recovery ———. 2009e. Building Resilient Communities: Risk Project.� Implementation Completion and Results Management and Response to Natural Disasters Report No. 26673, World Bank, Washington, DC. through Social Funds and Community-Driven ———. 2005a. “Malawi Emergency Drought Recovery Development Operations. Washington, DC: Project.� Implementation Completion and Results World Bank. http://siteresources.worldbank.org/ Report No. ICR0000520, World Bank, Washing- INTSF/Resources/Building_Resilient_Commun ton, DC. ities_Complete.pdf. ———. 2005b. “Mexico Natural Disaster Manage- ———. 2009f. “Samoa Cyclone Emergency Recovery ment Project.� Implementation Completion and Project.� Implementation Completion and Results Results Report No. 32168, World Bank, Washing- Report No. ICR00001071, World Bank, Washing- ton, DC. ton, DC. ———. 2006a. “Grenada Emergency Recovery and ———. 2009g. “Sri Lanka Tsunami Emergency Disaster Management Project.� Implementation Recovery Program.� Implementation Comple- 32 natural Disaster response: lessons from evaluations of the WorlD bank anD others tion and Results Report No. ICR00001105, World ———. 2010c. Safer Homes, Stronger Communities: Bank, Washington, DC. A Handbook for Reconstructing after Natural ———. 2010a. “Iran Bam Earthquake Emergency Disasters. Washington, DC: World Bank. Reconstruction Project.� Implementation Comple- ———. 2010d. “Uruguay Foot and Mouth Disease tion and Results Report No. ICR00001207, World Emergency Recovery Loan.� Implementation Bank, Washington, DC. Completion and Results Report No. ICR00001535, ———. 2010b. “Mali Fostering Agricultural Productiv- World Bank, Washington, DC. ity Project.� Project Appraisal Document, Report World Reconstruction Conference. 2011. “Proceed- No. 54332-ML, World Bank, Washington, DC. ings.� http://www.wrc-2011.org. 33