Resilience can be measured through the ability of people, communities, institutions, and systems to absorb shocks and recover. The figures below show both progress and continuing exposure: disaster mortality has fallen since 2005, while economic losses, climate risks, household pressures, and disruptions affecting children remain substantial. The periods differ by source, so historical observations, survey results, report estimates, and forecasts are labeled explicitly.
Contents
- Disaster losses and the return on resilience
- Climate exposure, poverty, and adaptation
- Economic and household resilience
- Physical, mental, and social well-being
- Children, conflict, and education
- Health-system climate resilience
Disaster losses and the return on resilience
The long-term cost of disasters has risen sharply. The Global Assessment Report (GAR) 2025 reports that average annual direct disaster costs were $70–80 billion during 1970–2000. During 2001–2020, the average rose to $180–200 billion per year. A broader 2025 report estimate that includes cascading and ecosystem costs puts current annual economic costs above $2.3 trillion.
The same GAR 2025 estimates describe resilience spending as an investment rather than only an expense. Each $1 spent on disaster-risk reduction delivers an average $15 return in averted future recovery costs. In sub-Saharan Africa, resilience and coping investments are estimated to produce long-term savings of up to 300% for droughts and up to 1,200% for storms. Yet only about 2% of global development assistance is allocated to disaster-risk reduction.
The benefits are not limited to financial loss. The UNDRR 2025 Annual Report says disaster-related mortality rates per 100,000 people fell by roughly 50% between 2005 and 2025. At the same time, 61 countries have experienced significant disaster-related fiscal shocks more than once per decade, according to a 2025 report estimate. Lower mortality therefore does not mean that public finances and communities face little recurring pressure.
Climate exposure, poverty, and adaptation
Climate resilience is closely tied to income and the ability to prepare. The World Bank’s Rising to the Challenge – Climate Adaptation and Resilience estimates that about 1.2 billion people worldwide are at high risk from climate-related hazards. It also finds that a 10% increase in income is associated with nearly 100 million fewer people at high risk. The assessment covers resilience interventions in 44 countries and reports that most are still lagging in implementation.
The poverty effects are already large. The World Bank’s Adaptation and Resilience: A Priority for Development and Poverty Reduction reports that global economic losses from climate-related disasters grew sevenfold from the 1970s to 2021. Natural disasters push about 26 million people into poverty every year, while climate change could push as many as 132 million people into poverty by 2030. A separate forecast in the same source says climate change and disasters could cause 216 million people to migrate within their own countries by 2050.
Infrastructure and warning systems offer measurable protection. In low- and middle-income countries, every $1 invested in resilient infrastructure yields an average $4 in benefits, and the estimated total net benefit over the lifetime of new assets is $4.2 trillion. Benefit-cost ratios for early-warning systems are greater than 4; ratios in other resilience sectors range from 2 to 10. These are report estimates, not guaranteed returns for every project.
| Resilience measure | Reported result | Period or status |
|---|---|---|
| Annual direct disaster costs | $70–80 billion average | 1970–2000 |
| Annual direct disaster costs | $180–200 billion average | 2001–2020 |
| Return from each $1 in disaster-risk reduction | $15 average | 2025 report estimate |
| Return from each $1 in resilient infrastructure | $4 average | 2021 report estimate |
| People at high climate-hazard risk | About 1.2 billion | 2024 report estimate |
Economic and household resilience
Labor markets in OECD countries recovered strongly after the initial COVID-19 shock, but household security remains uneven. According to How’s Life? 2024, the average OECD employment rate for people aged 25–64 was 78.4% in 2023, 1.6 percentage points above its pre-COVID level. The OECD unemployment rate reached a record-low 4.9% in December 2023. Average employment rates had returned to historical highs by the first quarter of 2024 after dropping during the first six months of 2020.
Household pressures tell a less uniform story. Since 2019, the share of low-income households overburdened by housing costs has increased in one-third of OECD countries. In European OECD countries, one in 11 people could not afford to keep their home adequately warm in 2023, compared with one in 14 in 2019. Just under 9% of people in 17 European OECD countries were unable to keep their home adequately warm in 2023.
Before the pandemic, the share of people reporting difficulty making ends meet fell from 30% to 19% across OECD countries over the preceding decade. Between 2019 and 2023, however, the OECD-average share reporting this difficulty rose by 0.03 percentage points per year. Average household debt was 120% of household net disposable income in 2022. In the same year, government financial liabilities exceeded financial assets by 20 percentage points of GDP on average across OECD countries.
Together, these figures distinguish employment recovery from broader financial resilience. A person may be employed while facing housing-cost, energy, or debt pressure, and a country may restore its employment rate while public balance sheets remain exposed.
Physical, mental, and social well-being
Resilience also includes the capacity to maintain health and social connection during prolonged disruption. The OECD reports that average life expectancy across OECD countries fell by almost half a year during the COVID-19 period through 2023. Almost 30% of people in OECD countries reported experiencing a lot of physical pain in 2022–23. The prevalence of physical pain rose by an annualized 0.4 percentage points between 2017–19 and 2022–23, and the share increased by at least 3 percentage points in two-fifths of OECD countries relative to 2017–19.
Emotional indicators also worsened in the latest period. OECD-average sadness rose by 0.3 percentage points per year from 2017–19 to 2022–23, compared with 0.06 points per year from 2008–10 to 2017–19. Feelings of sadness stagnated in 24 of 38 OECD countries and deteriorated in 10 during the latest four-year period, using 2019–2023 or the latest available data.
Social contact declined in the available time-use comparisons. Around 2018, people across OECD countries spent about six hours per week interacting with friends and family. In Canada, Japan, and the United States, average weekly social-interaction time fell by 40 minutes between 2009–11 and 2015–22.
Institutional trust provides another context for collective resilience. In the late-2023 OECD Trust Survey, 44% reported low or no trust in national government, while 39% reported high or moderately high trust. These OECD figures are averages and country-count summaries; they should not be interpreted as direct measures of any individual person’s psychological resilience.
Children, conflict, and education
Children face overlapping shocks from conflict, climate hazards, and disrupted schooling. The UNICEF Prospects for Children in 2025 report says more than 473 million children lived in areas affected by conflict in 2023. Children affected by conflict represented almost 19% of children globally, up from around 10% in the 1990s.
The UNICEF Children’s Climate Risk Report 2026 estimates that nearly half of the world’s children—1.1 billion—are at risk from at least three overlapping climate hazards. More than 4 million children face as many as six overlapping climate threats. These are 2026 report estimates, not measurements of one single event or a forecast unless specifically labeled as one.
Education systems are another measure of resilience. UNESCO’s Crisis-sensitive educational planning page gives a 2026 estimate of about 224 million school-aged children affected by crises globally, including about 7 million school-aged refugees deprived of their right to education. The same page estimates that about 1 billion children face extremely high risk from climate-change impacts. These are the latest estimates displayed on pages updated in 2026, rather than forecasts.
The scale of these figures shows why child resilience depends on systems around the child: safety, continuity of education, climate adaptation, and support for displaced families. Conflict and climate exposure can overlap, so the categories should not be added together as if they represented separate groups.
Health-system climate resilience
National adaptation planning shows a gap between recognizing health and planning specific system action. The World Health Organization’s Building Climate-Resilient Health Systems page reports that health was identified as a priority sector by 100% of countries that had submitted National Adaptation Plan documents as of December 2020.
However, only around 50% of submitted plans specifically identified impacts on health systems as an adverse effect of climate change. Among assessed plans, 89% included some assessment of health-related climate vulnerabilities, while 40% included further health-system vulnerability assessment as a planned action. The measures therefore range from broad recognition to a concrete planned assessment, and those are not equivalent stages of implementation.
Health systems also have an environmental footprint. The WHO Europe page on climate-resilient and environmentally friendly health systems estimates that the health-care sector accounts for more than 4.4% of global net climate emissions. Building resilience consequently has two linked dimensions: maintaining care during climate stress and reducing the sector’s contribution to the conditions that intensify that stress.
These statistics describe sector-specific resilience across different populations and periods. They are descriptive figures from official or intergovernmental sources, not advice, and the definitions should remain attached to each measure when comparing disaster risk, household security, well-being, children’s exposure, education access, or health-system preparedness.