Disasters caused by natural hazards are increasingly threatening the lives and livelihoods of the world’s poor and disaster-vulnerable populations. Disasters resulting from natural hazards cause billions of dollars in damages per year, and these costs are rising due to population growth, rapid and unplanned urbanization, low-quality infrastructure, and ineffective disaster risk governance. Some 82 percent of deaths since 1970 caused by natural hazards and extreme weather occurred in low- and lower-middle-income countries, where most people living in poverty reside (World Bank 2020c). Climate change is further exacerbating negative disaster impacts by contributing to more destructive and frequent droughts, floods, and storms. These disasters disproportionately affect poor populations, especially disadvantaged or vulnerable populations, who typically reside in areas with high exposure, lose larger shares of their wealth to disasters, and have limited assets and access to social networks for coping and recovery.

This evaluation focuses on the World Bank’s support for reducing disaster risks caused by natural hazards. Disaster risk reduction (DRR), as defined by the United Nations Office for Disaster Risk Reduction, “aims to prevent new—and reduce existing—disaster risk while managing residual risk, all of which contribute to strengthening resilience to achieve sustainable development” (UNDRR 2016). According to the Sendai Framework for Disaster Risk Reduction 2015–2030, DRR practices are multihazard and multisectoral: they address disaster risks caused by natural, environmental, technological, and biological hazards. This evaluation focuses on one type of DRR—that is, efforts to reduce disaster risks caused by natural hazards. To assess these efforts, the evaluation uses universally accepted terminology associated with efforts to minimize exposure and vulnerability, as well as other activities that can limit negative disaster effects. “Exposure” is the situation of people, infrastructure, housing, production capacities, and other tangible and intangible (for example, cultural) assets located in hazard areas (UNDRR 2016). “Vulnerability” refers to the conditions determined by physical, social, economic, and environmental factors that increase the susceptibility of an individual, a community, assets, or systems to the impacts of hazards (UNDRR 2016).

Investing in DRR has strong economic and social benefits, yet underinvestment in DRR globally—particularly in disaster risk mitigation and preparedness—remains an issue. Resilient infrastructure investments can have a present value of $4 return on each dollar invested (Hallegatte, Rentschler, and Rozenberg 2019). When countries rebuild infrastructure after disasters to be more resilient, they can reduce the negative impact of future disasters on well-being by as much as 31 percent (Hallegatte, Rentschler, and Walsh 2018). Universal access to an early-warning system (EWS) can reduce asset and well-being losses from disasters by an estimated $35 billion per year. An EWS can also contribute to a decrease in mortality (Hallegatte et al. 2017). Yet only 4.1 percent of total official development assistance for disasters was directed toward disaster prevention and preparedness between 2010 and 2019 (UNDRR 2021). This is due to insufficient resources for investment at the country level, limited knowledge of disaster risks and vulnerabilities, and existing government preferences for politically visible post-disaster initiatives rather than pre-DRR measures.

Purpose and Scope of the Evaluation

The purpose of this evaluation is to surface lessons on the World Bank’s support for DRR. The evaluation focused on World Bank support to address disaster risks caused by natural hazards, not on other types of hazards or stresses. It covers disaster risks caused by floods, droughts, earthquakes, cyclones, volcanic disruptions, tsunamis, and landslides. Areas of DRR support covered include risk identification; risk reduction activities, such as resilient infrastructure, buildings, and protective works; the integration of DRR into institutions, policy, and planning; preparedness activities, including support for EWSs; and disaster risk finance. The evaluation does not cover elements of disaster response and recovery that do not have DRR activities. The evaluation does not cover the International Finance Corporation or the Multilateral Investment Guarantee Agency because DRR is not a major corporate priority for either organization.

Main Findings

The World Bank has approved a large and growing portfolio of DRR activities that help clients mitigate, prepare for, and recover from disasters caused by natural hazards. The World Bank has tripled its DRR support since fiscal year 2010, approving 634 lending operations and 504 nonlending products with DRR activities, and it has expanded DRR lending operations into many countries where this type of lending was historically infrequent. The growth of DRR support is associated with the World Bank’s corporate prioritization of climate change adaptation, the special theme on climate change of the International Development Association (IDA), the presence of the Global Facility for Disaster Reduction and Recovery, and a supportive global authorizing environment.

Strategic Alignment

The World Bank’s support for DRR has been highly relevant. It focuses its DRR work on those countries with the most serious natural hazards. It often uses multiple and synergistic pillars of DRR engagement that include hazard identification, resilient infrastructure, early-warning and preparedness activities, and disaster risk finance on occasion. The World Bank has also shifted its focus from post-disaster response toward risk reduction, which it has built into nearly all disaster response activities. The share of World Bank DRR projects that engage in predisaster, as compared with ex post, steadily rose from 50 percent in fiscal year 2010 to 80 percent in fiscal year 2020, whereas the share of projects supporting disaster response without DRR elements declined from 30 percent to 0 percent between 2010 and 2020.

The World Bank has made significant progress in mainstreaming DRR in lending operations, but there has been less uptake in some sectors. The number of mainstreamed DRR projects has quadrupled over the evaluation period, with growth occurring across most relevant Global Practices (GPs). However, many GPs are starting from a low base: the share of all approved operations that include DRR activities remains low in Agriculture and Food and Energy and Extractives, and mainstreaming does not happen evenly across relevant subsectors. These gaps may be because of the tendency of the World Bank to focus its DRR activities on exposed infrastructure in urban areas while it has less familiarity with or emphasis on DRR in other sectors.

World Bank support for DRR in IDA countries, small island developing states, and low-income countries experiencing fragility, conflict, and violence (FCV) has been particularly comprehensive. The World Bank has provided at least some lending support for DRR in almost all IDA countries facing the most serious hazards and has provided lending support for all of the most serious hazards in more than 60 percent of IDA countries. There is also even more comprehensive coverage of nonlending work in IDA countries. Similarly, the World Bank has provided DRR lending support for 95 percent of small island developing states that face the most serious hazards and 85 percent of countries experiencing FCV as well as the most serious hazards, most of which are eligible for IDA assistance.

However, there are coverage gaps in the Middle East and North Africa and in Europe and Central Asia and varying levels of coverage across hazard types. In these Regions, DRR lending for serious hazards is more often lacking, as there are fewer lending projects with DRR activities, infrastructure operations are less likely to include mainstreamed DRR, and DRR support is less integrated. Coverage gaps are associated with the prevalence of International Bank for Reconstruction and Development countries that cite the high direct costs and opportunity costs of borrowing for DRR, the presence of fragility and conflict that impede DRR uptake, financing decisions to prioritize high-intensity and frequently occurring hazards, and other external factors such as European Union directives that focus on floods but not other hazards. Some hazard types that are rarer (tsunamis, volcanic eruptions) or less catastrophic (landslides) receive less attention than others (floods, cyclones, droughts, and earthquakes) in World Bank engagements.

Building Country Engagement

The World Bank has overcome constraints to DRR client uptake by persuading the right decision makers using evidence-informed engagements and by engaging in disaster reconstruction or sector programs. Rigorous analytical work that quantifies risks, assesses costs and benefits, and communicates impacts has influenced clients to undertake DRR actions. Many of these analyses were funded by the Global Facility for Disaster Reduction and Recovery. Working with committed government actors with decision-making power was a key factor: when the World Bank mainly worked with disaster agencies, progress on DRR was slow. The World Bank’s support for disaster reconstruction has been an important entry point for engaging on DRR, as has the credibility earned from sustained sector engagements.

Effectiveness of World Bank Disaster Risk Reduction Activities

The World Bank is often not able to demonstrate the effects of its DRR activities on reduced exposure and vulnerability, which has consequences for its ability to make a development case for risk reduction. Most DRR operations do not provide sufficient information to establish the level of DRR being achieved. This inhibits the World Bank’s understanding of the level at which DRR contributes to development impacts, such as reduced economic loss and mortality. This lack of information is especially apparent for resilient infrastructure investments and development policy operations. Most resilient infrastructure investment projects lack information on resilience standards. Many development policy operations lack evidence on the results of policy changes.

The World Bank is increasingly identifying and addressing the needs of some groups that are disproportionately impacted by disasters; however, for other groups, there is slow progress and limited reporting on DRR benefits. There is more meaningful coverage of gender-DRR issues, but few operations integrate the needs of other identified disaster-vulnerable groups, including persons with disabilities, the elderly, children, and youth.

Although DRR engagements in conflict-affected situations have addressed disaster vulnerability, they have missed opportunities to use conflict-sensitive approaches to mitigate conflict risks and to pursue peace-building opportunities. Conflict can be a key driver of disaster risk, and disaster risk may exacerbate preexisting conflicts and increase the risk of violence. Yet there is no established methodology for conducting a hybrid Post-Disaster Needs Assessment that integrates a conflict lens, and efforts to develop a DRR-FCV program in the World Bank have been stymied by a lack of donor support.

The evaluation assessed results and generated lessons on factors of effectiveness for four key approaches that have been reported on by the World Bank in its periodic updates to the Board of Executive Directors. These approaches were (i) disaster-resilient infrastructure, (ii) EWSs, (iii) disaster insurance, and (iv) DRR policy reforms. The evaluation found the following:

• DRR investment projects often build effective relevant infrastructure, but most of these projects do not explicitly address operations and maintenance that are required for long-term resilience aims. This shortcoming is more evident in core disaster projects mapped to the Urban, Disaster Risk Management, Resilience, and Land GP, as compared with sectoral infrastructure projects.
• The World Bank has been more effective in developing EWS infrastructure than in delivering EWS services, such as forecasting capacity and community-
  preparedness activities.
• Disaster insurance activities have had a limited impact on transferring disaster risk because insurance programs have had difficulty in reaching scale. However, disaster insurance activities have made progress in raising awareness, building capacity, and developing products.
• Although development policy financing projects with DRR policy actions have mostly achieved their disaster-related indicators, they often have not demonstrated downstream impacts or changes in disaster-related behaviors in the real economy.

The World Bank has been able to achieve highly successful results on DRR through flagship programs where it brought the full weight of the institution to bear using sustained engagement, prioritization in policy dialogue, sizable lending programs, access to trust funds, and catalyzation of financing from others. By necessity, the World Bank can do this for only a limited number of cases at a time, so it must consider when its involvement in a program has been sufficient and when to change course to tackle the next difficult problem where it has a comparative advantage.

Recommendations

Recommendation 1. Incorporate DRR activities in regions and sectors and for hazards that exhibit significant coverage gaps. In countries facing high risks from disasters caused by natural hazards, the World Bank can address coverage gaps through analytical work, mainstreaming, or core DRR activities, including by (i) conducting country-level analytics on disaster costs and impacts of DRR for key sectors, (ii) relying on country management to proactively engage clients on DRR and encourage task teams to integrate DRR considerations in projects, (iii) integrating DRR specialists into sector dialogue, and (iv) assessing the need for coverage of low-frequency but catastrophic hazards such as volcanic eruptions and tsunamis.

Recommendation 2. Identify and measure the effects of DRR activities on exposure and vulnerability to strengthen the development case to clients facing serious disaster risks. The generation of ex post DRR evidence on probable outcomes involves clearer articulation in project documents of the particular resilience standards used for infrastructure in that context, use of and reporting on verification mechanisms for compliance with these standards, and greater use of ex post modeling of the incremental impacts of DRR activities on expected damage, loss, and mortality from disasters. This evidence generation can occur in projects or from results assessments of DRR activities implemented in different contexts.

Recommendation 3. Integrate the needs of populations that are disproportionately vulnerable to disasters caused by natural hazards into DRR project targeting and design, implementation, and results reporting. This can be accomplished by strengthening collaboration between the GPs working on disaster activities with poverty and social development experts in the World Bank through the development and application of data, tools, analyses, and tracking systems.

Recommendation 4. In countries affected by serious natural hazards and fragility and conflict risks, identify and assess the ways in which hazards and conflict interrelate, and use this knowledge to inform country engagement and project design. This should form a part of the way the World Bank is increasingly addressing compound risks at the country level. Taking such steps may require strengthened collaboration and knowledge exchange between World Bank DRR and FCV teams, the use of integrated multirisk analysis tools, and adapted program designs that address the interlinkages between disaster and FCV risks.