Kenya

Current socio-economic context and recovery 

Kenya’s population of 51 million is predominantly rural, with only 30 per cent living in cities. Multidimensional poverty affects 44.7 per cent of the population. This vulnerability is reinforced by broader structural factors. A large share of the population is young, with over a third under the age of 18. Gender inequality remains significant, shaping unequal access to income, assets, and opportunities. Persons with disabilities and other vulnerable groups face additional barriers to accessing support and services. At the same time, early warning systems remain limited, reducing the ability of communities to anticipate and respond to hazards effectively.

Source note: Indicator-level sources are shown in the graphic. For full metadata, including definitions, units, time coverage and source links, see the methodology.

Geographic distribution of hotspots for key hazards across the country

Extreme drought in northern areas is one of the main drivers of economic losses in Kenya. This risk is amplified by frequent and destructive flooding in the Lake Victoria Basin, along major river systems, and in fast-growing urban centres such as Nairobi. In the Central Highlands and other steep, densely populated areas, intense rainfall also creates persistent landslide risk. Heatwave risk adds another layer to this geography of risk, with higher annual probabilities particularly visible along the coast and other low-lying areas.

Average Annual Losses by hazards

In economic terms, flood risk is the main driver of expected direct losses, accounting for more than 60% of total impacts. This would cost $125 million of a predicted $200 million in total losses.This reflects both the frequency of flooding and its wide geographic reach across the country. Earthquakes, while less visible in everyday risk discussions, represent the second largest source of direct losses at around 30%, highlighting their potential for high-impact events. Landslides, despite being locally important in highland regions, contribute only a very small share of total direct losses (around 1%) at the national level.

Note: Drought is not included in this AAL distribution, but remains central to Kenya’s risk profile given the importance of agriculture for livelihoods, employment and climate exposure. Its impacts are therefore presented separately through crop-loss and macro-fiscal evidence.

Note: CDRI/GIRI AAL estimates cover modelled direct damages to buildings and infrastructure for selected hazards only. They do not yet provide comparable AAL estimates for all relevant hazards, including agricultural drought, heatwaves and wildfires. Figures should therefore be interpreted within the scope of the CDRI/GIRI methodology and not directly compared with estimates from other sources. UNDRR’s Risk and Resilience Metrics Facility aims to progressively develop methodologies to integrate these additional hazards into future comparable risk metrics.

Direct average annual losses (AAL) to infrastructure. (Costliest hazard)

Losses from floods are concentrated in buildings and key public-service infrastructure. Buildings account for the largest share of modelled losses, at around US$85.1 million, followed by education assets at US$21.4 million and power infrastructure at US$8.3 million. Telecommunications, roads and railways, water and wastewater, health facilities, ports and airports, and oil and gas assets account for smaller shares.

Note: Information presented for the CDRI/GIRI top hazard's AAL estimation.

Direct average annual losses (AAL) to infrastructure. (Second most costly hazard)

Earthquake losses are also concentrated in buildings and public-service infrastructure. Buildings account for the largest share of modelled losses, at around US$37.6 million, followed by education assets at US$9.4 million, telecommunications at US$4.1 million, roads and railways at US$3.8 million, water and wastewater at US$2.1 million, and power infrastructure at US$2.1 million. Health, ports and airports, and oil and gas assets account for smaller shares.

Note: Information presented for the CDRI/GIRI top hazard's AAL estimation.

Expected economic losses in 1-in-100 year event (PML)

Kenya’s selected extreme-loss estimates show that drought and flood risks could generate the largest losses among the hazards and sectors presented. However, these values should be interpreted by source and scope. Flood, tsunami and landslide estimates refer to CDRI/GIRI modelled direct damages to buildings and infrastructure, while drought and ecosystem-related estimates come from separate sectoral analyses. The chart therefore highlights the breadth of Kenya’s exposure across built assets, agriculture and natural systems, rather than providing a fully harmonised ranking of all hazards

Direct Average Annual Losses to public infrastructure

In the education sector, wind has the highest estimated losses, at around US$16.04 million per year, followed by storm surge at US$15.59 million, earthquakes at US$9.65 million, and floods at US$4.58 million. Health infrastructure losses are lower across all hazards, with estimated annual losses of around US$109,082 from wind, US$106,072 from storm surge, US$65,673 from earthquakes, and US$31,148 from floods.

Direct Probable Maximum Losses to public infrastructure

The largest sector-level estimate is associated with flood impacts on power infrastructure, at approximately US$55.9 million. Earthquake-related losses are highest for telecommunications (US$53.0 million) and roads and railways (US$49.3 million), followed by power and water and wastewater infrastructure, both at around US$27 million. For floods, the largest estimates after power are recorded for telecommunications (US$35.8 million) and roads and railways (US$27.0 million). Rain-related losses are reported only for roads and railways, at US$20.6 million. Tsunami-related losses are lower across the sectors shown, with telecommunications recording the highest estimate at US$9.4 million. Oil and gas, and ports and airports, record comparatively low PML estimates across the modelled hazards.

Direct Average Annual Losses to housing

The largest losses are reported for the low-middle income class across all hazards, with around US$26.02 million from floods, US$11.45 million from earthquakes, and US$2.80 million from tsunamis. The low income class records estimated losses of around US$4.31 million from floods, US$1.90 million from earthquakes, and US$0.50 million from tsunamis. For high income and middle income classes no values are reported.

Probable Maximum Losses of real GDP now and 2050

In contrast to Kenya’s high vulnerability to climate and disaster-related risks, the model shows limited impacts from flood and earthquakes on the economy and fiscal trajectories. The main reason is the model considers only the direct damages to built physical assets (buildings and infrastructure), which does not include crops, food yields, and other components in the agriculture sector which are vital to Kenya’s economy. Putting this limitation aside, the damages to buildings and infrastructure result in only limited impacts to the country’s real GDP growth, with no scenario resulting more than -0.4% in decline. The growth decline ranges from -0.04% for an AAL and between -0.1 to -0.4% for PML flood scenarios, while earthquake events decrease growth by -0.02 for AAL and between -0.03 and -0.3% for PML scenarios. The main sources of damages come from commercial and industrial buildings and critical infrastructure including the power (floods) and telecommunication (earthquakes) sectors.

Chance of a disaster exceeding public financing capacity (Fiscal gap)

Results show that the modelled disaster scenarios have limited impacts on Kenya’s public debt. The PML scenarios are expected to increase the debt ratio only marginally, from 66.4% of GDP to between 66.5% and 66.7% for floods, and from 66.4% of GDP to 66.7% for earthquakes. However, the projections also show that debt continues to rise over time after the shock, as public reconstruction costs and existing debt-servicing costs place additional pressure on government finances.

Sovereign debt due to an extreme event

The modelled 1-in-100-year loss is equivalent to around 0.15% of Kenya’s 2025 GDP. In fiscal terms, this represents approximately 0.83% of annual government revenue, 0.61% of public expenditure and 0.21% of gross public debt. While this suggests a limited direct loss at aggregate macro-fiscal level, Kenya’s fiscal position is already constrained, with a fiscal deficit of 6.41% of GDP, a primary deficit of 1.06% of GDP and gross public debt of 69.33% of GDP. Reconstruction needs, indirect losses or contingent liabilities could therefore place additional pressure on fiscal space.

These figures are scale comparisons and do not estimate the resulting increase in public debt or the government’s actual contingent liability.

Disaster losses to assets and consumption

The welfare results translate asset losses into human impact. The chart compares asset risk and well-being risk under current conditions and under a modelled resilience scenario. Asset risk falls from 0.22% to 0.19% of GDP, while well-being risk falls from 0.47% to 0.29% of GDP.

Household consumption resilience after a severe shock

Kenya’s current household consumption resilience – the economy’s capacity to absorb asset losses before they translate into welfare losses – stands at just 46 per cent. With resilience investment, household consumption resilience rises to 68 per cent, a 22-percentage-point improvement.

Estimated cost, benefits and investment needs: agriculture and resilience (2031–2050)

The study estimates that the projected benefits of resilient, low-carbon agriculture are almost six times larger than the investment amount assessed, and around fifteen times higher than the estimated economic cost.

Estimated cost, benefits and investment needs: Forest landscapes (2031–2050)

The low-carbon and resilient climate scenario has higher estimated costs and investment needs than the business-as-usual scenario, but also higher estimated benefits. In the low-carbon and resilient climate scenario, combined benefits from timber revenue and reduced externalities are estimated at around US$9.38 billion, compared with combined investment needs and economic costs of around US$5.32 billion. Under business as usual, combined benefits are estimated at around US$461 million, compared with combined investment needs and economic costs of around US$244 million. Results should be interpreted within the scope and assumptions of the underlying study.