## 1. Current Health Exposure to Risks Exacerbated by Climate Change

## Source details

**Canonical URL:** [1. Current Health Exposure to Risks Exacerbated by Climate Change](https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024049-print-pdf.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2024/english/wpiea2024049-print-pdf.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2024/english/wpiea2024049-print-pdf.pdf.json)

---

### Background and framing
- Climate change produces direct and indirect effects on health:
  - Direct: mortality and morbidity from extreme heat and weather events such as storms, forest fires, floods, or droughts.
  - Indirect: changes mediated through ecosystems (disease patterns, agriculture), economic activity (air pollution), and social structure (migration and conflict).
- Vulnerable groups disproportionately affected include the elderly, children, and those with pre-existing health conditions.
- Adaptation of health systems requires a comprehensive approach spanning prevention, preparedness, response, and recovery:
  - Prevention (largely outside the health sector) includes greenhouse gas emissions reductions.
  - Preparedness: enhance public health infrastructure, surveillance, and early warning systems.
  - Response: timely emergency medical care during extreme events.
  - Recovery: rebuilding health systems after realized climate-related damage.
- Urgency emphasized: achieve SDG 3 by 2030 while adapting health systems to climate change.

### Key contextual estimates and comparative findings
- Adapting infrastructures to climate change: global net present value of over $2.5 trillion (Hallegatte et al., 2019b).
- Energy share of greenhouse gas emissions: 34 percent (Lamb et al., 2021).
- Health care systems emissions: 4.4 percent of emissions; 17 percent of these emissions are in Scope 1, implying 0.7 percent overall in Scope 1 (Setoguchi, Leddin, Metz and Omary, 2022).
- Study coverage: results across five sectors (health, education, water and sanitation, electricity, roads) and 173 countries.
- Aggregate across five sectors including health:
  - Additional 3.8 percent of global GDP, or US$3.4 trillion, of public and private spending required by 2030 to achieve a strong performance in the selected SDGs while addressing associated climate risks.
  - Includes a sizeable annualized increase of 0.4 percent of global GDP (US$358 billion) in 2030 compared to estimates that do not account for mitigation and adaptation needs within these sectors.

### Methodology overview for health (SDG 3)
- Empirical approach: benchmarking analogous to Gaspar et al. (2019) to estimate expenditure needed to adapt health systems to climate impacts.
- Three main components:
  1. Identify countries’ current and future exposure to climate change (Section 1).
  2. Assess countries’ performance in reducing vulnerability given this exposure (Section 2).
  3. Estimate additional cost to achieve “good performance” by benchmarking countries to high performers among peers with similar exposure and projecting needs based on future exposure levels (Section 3).
- Focus: adaptation spending in health (no quantified mitigation spending in SDG 3 due to absence of well-established sectoral mitigation standards comparable to education).
- Output metric: additional per-capita health adaptation costs expressed as a percent of 2030 GDP.

### Measurement framework — three principal dimensions
- Dimensions (IPCC (2014) and WHO (2015)):
  - (i) health impacts of climate-change-exacerbated extreme weather and disaster events;
  - (ii) exposure to climate-sensitive parasitic and viral diseases;
  - (iii) exposure to the risk of malnutrition brought about by climate change.
- Overall exposure constructed across m=8 exposure types: heatwaves, wildfires, cyclones, tsunamis, floods, droughts, malaria, arboviral diseases, food scarcity.
- Each indicator rescaled to range 0–10 via min-max normalization and aggregated with weights w_e equal to the proportion of deaths D attributable to each exposure type:
  - w_e = D_e / sum_{e=1}^m D_e
- Countries classified into five groups with an equal number of countries: very low, low, medium, high, and very high.

### a. Health exposure to climate-change-induced extreme weather and disaster events (indicators and methods)
- Heatwaves:
  - Indicator: change in the number of heatwave exposure events (one exposure event = one heatwave experienced by one person aged over 65 or under 1 year old).
  - Baseline comparison: change from latest year (2020) measured with respect to cross-year average in reference period (1986–2005).
  - Heatwave definition: at least two days during which both daily minimum and maximum temperatures are above the 95th percentile of their respective climatologies.
- Wildfire:
  - Uses model-based wildfire danger and satellite-observed exposure.
  - Climatological wildfire danger: daily ‘very high’ or ‘extremely high’ danger (fire danger index score 5 or 6 on a range of 1-6) combined with climate and population data.
  - Human exposure measured in person-days, averaged over 2017-2020.
- Droughts:
  - Agricultural drought: dry period where at least 30 percent of the crop area was in stress for more than 10 days.
  - Measured using Agriculture Stress Index (ASI) based on Vegetation Health Index (VHI); percent pixels in arable areas with VHI value below 35 percent used.
  - Drought probability: frequency of droughts within the last 30 years.
  - Historical affected population: number of historically affected people per year based on last 25 years.
  - Final drought exposure: mean between average annual size of drought-affected population and frequency of drought events.
- Floods, tsunamis, tropical cyclones:
  - Hazard zones from hazard maps and event return-periods overlaid with population distribution to derive population exposed by return-period.
- Population exposure metric:
  - Mean of absolute (unnormalized number of exposed people) and relative (share of total population exposed) population exposure used after applying a log transformation to absolute exposed people.
  - Min-max normalization formula preserved exactly as specified in source.

### b. Exposure to climate-sensitive parasitic and viral diseases
- R0 of Aedes aegypti and Aedes albopictus:
  - Measures environmental suitability for arbovirus transmission (dengue, chikungunya, Zika).
  - Based on model capturing temperature and rainfall effects on vectorial capacity and abundance; overlaid with population density to estimate R0.
  - Variable used: average of indicators for these two vectors.
- Malaria transmission season length:
  - Threshold-based model incorporating precipitation accumulation, average temperature, and relative humidity to estimate length of transmission season for Plasmodium falciparum.

### c. Exposure to risk of malnutrition attributable to climate change
- Indicator: change in crop growth duration (time to reach a target sum of accumulated temperatures, ATT) for maize, winter wheat, spring wheat, rice, and soybean, measured against 1981-2010 reference period.
  - Crop growth duration measured by 11-year rolling average of days to reach ATT.
  - Negative change in ATT relative to reference implies shortened growth duration, more rapid maturing, smaller crops, and lower yield.
- Composite weighted-average indicator:
  - Weights w_c for crop c reflect importance based on value of production Y:
    - w_c = Y_c / sum_{c=1}^n Y_c
- Outliers removed using Tukey rule:
  - x_i < Q1 − 1.5·IQR or x_i > Q3 + 1.5·IQR
- Min-max normalization applied to non-outlier observations.

### d. Data sources and normalization
- Data sources:
  - INFORM Risk database for tsunamis, floods, tropical cyclones, drought, and wildfire.
  - Lancet Countdown for heatwaves, vector-borne suitability, and malnutrition exposure indicators.
- Normalization: each indicator normalized to 0–10, then weighted-average computed to produce overall exposure metric.

### e. Future projected health exposure to climate change (projections and assumptions)
- Scenario and indicators substituted with ND-GAIN under RCP 4.5 (“moderate emissions scenario”).
- Heatwaves: absolute change in Warm Spell Duration Index (WSDI) from baseline (1960-1990) to future (2040-2070). WSDI counts days where daily maximum near-surface temperature exceeds 90th percentile threshold for six or more consecutive days.
- Floods: percentage change in flood hazard from baseline (1960-1990) to future (2040-2070) based on predicted monthly maximum precipitation in five consecutive days.
- Proxies: current exposure measures used for tsunamis, tropical cyclones, droughts, and wildfire when future projection data unavailable.
- Vector-borne diseases: absolute change in malaria Length Transmission Season (LTS) from baseline (1980-2010) to 2050 projection.
- Malnutrition: forecasts from five crop models for rice, wheat, maize; projected change as percentage change from baseline (1980-2009) to future (2040 to 2069).
- Projected overall metric: normalize indicators to 0–10 and classify countries into five equal-sized exposure groups.

### 2. Performance in Reducing the Vulnerability of Health Systems to Climate Risks
- Purpose: construct measures of countries’ performance in reducing vulnerability to climate risks to health; indicators reflect use of resources to reduce populations’ vulnerability.
- Four performance indices (normalized analogously to exposure construction):
  - Heat-related mortality:
    - Captures heat-related mortality in populations older than 65 years using exposure-response function and optimum temperature from Honda et al. (2014).
  - Lethality of natural hazards (wildfires, floods, landslides, droughts, storms):
    - Sums total deaths per country arising from each disaster.
    - Event considered a disaster if at least one criterion met:
      - (i) 10 or more people die,
      - (ii) 100 or more people are affected,
      - (iii) declaration of a state of emergency,
      - (iv) call for international assistance.
  - Climate-sensitive parasitic or viral diseases:
    - Indicator is sum of deaths due to dengue and lymphatic filariasis, and total malaria cases.
  - Malnutrition attributable to climate change:
    - Measure used is child stunting: number of under-five-year-olds falling below −2 standard deviations (moderate and severe stunting defined as below −2 and −3).
- Composite performance variable:
  - Weighted-average with weights w_p equal to share of deaths from each of n event types p:
    - w_p = D_p / sum_{p=1}^n D_p
- High performer definition:
  - Country considered a high performer if normalized score > 9.
  - If more than ten countries in an exposure group meet this, only the ten highest performers in that group are considered.

### 3. Additional expenditure to adapt health systems to climate risks (benchmarking and costing)
- Lancet Countdown tracks spending on health adaptation by country, including public and private spending on health infrastructure development, disease surveillance, disease control systems, health workforce training, and key IPCC-identified adaptation measures (Romanello et al., 2022).
- Benchmarking approach:
  - Country i’s peer group = projected exposure group to which it belongs.
  - Benchmark against ‘good performers’—those most effective in reducing population health systems’ exposure given current exposure and vulnerability.
  - Additional expenditure for country i = difference between mean per-capita health adaptation costs of well-performing peers and country i’s per-capita health adaptation costs.
  - Results reported as percent of 2030 GDP.
- Scope: public and private spending on adaptation measures for health systems as defined above.
- Benchmarking uses projected exposure-group membership and current exposure and vulnerability measures.

### Policy relevance, limitations, and intended use
- Results intended to inform national policymakers and multilateral stakeholders and to support integration into National Adaptation Plans (NAPs), Nationally Determined Contributions (NDCs), and SDG-aligned budgeting processes.
- Limitations and uncertainties explicitly noted:
  - Scientific understanding limits, data quality and availability constraints, model construction choices, and reliance on expert judgment where needed.
  - Uncertainties about availability and costs of future technologies and demographic and socio-economic trajectories.
- Framing:
  - Paper accounts for both adaptation and mitigation where relevant; for health the focus is on adaptation.
  - Findings meant to guide resource allocation, highlight financing needs, and motivate international cooperation.

*Source: IMF Working Paper — "Accounting for climate risks in costing the Sustainable Development Goals", wpiea2024049-print-pdf (Section 1 and related excerpts).*

### 1. Current Health Exposure to Risks Exacerbated by Climate Change .................................................. 9

### 1. Current Health Exposure to Risks Exacerbated by Climate Change

### Background: health impacts of climate change
- Climate change produces direct and indirect effects on health:
  - Direct: mortality and morbidity from extreme heat and weather events such as storms, forest fires, floods, or droughts.
  - Indirect: changes mediated through ecosystems (disease patterns, agriculture), economic activity (air pollution), and social structure (migration and conflict).
- Vulnerable groups disproportionately affected include the elderly, children, and those with pre-existing health conditions.
- Adaptation of health systems requires a comprehensive approach spanning prevention, preparedness, response, and recovery:
  - Prevention (largely outside the health sector) includes greenhouse gas emissions reductions.
  - Preparedness: enhance public health infrastructure, surveillance, and early warning systems.
  - Response: timely emergency medical care during extreme events.
  - Recovery: rebuilding health systems after realized climate-related damage.
- The paper emphasizes the urgency of achieving SDG 3 by 2030 while adapting health systems to climate change.

### Key contextual estimates and comparative findings cited
- Adapting infrastructures to climate change is highly cost effective, with a global net present value of over $2.5 trillion (Hallegatte et al., 2019b).
- Energy accounts for 34 percent of total greenhouse gas emissions (Lamb et al., 2021).
- Health care systems are responsible for 4.4 percent of emissions; 17 percent of these emissions are in Scope 1, implying 0.7 percent overall in Scope 1 (Setoguchi, Leddin, Metz and Omary, 2022).
- The study produces global results across five sectors (health, education, water and sanitation, electricity, roads) and 173 countries (country coverage referenced in Table A1 in the Appendix).
- Aggregate finding across the five sectors (including health): an additional 3.8 percent of global GDP, or US$3.4 trillion, of public and private spending will be required by 2030 to achieve a strong performance in the selected SDGs while addressing associated climate risks.
  - This includes a sizeable annualized increase of 0.4 percent of global GDP (US$358 billion) in 2030 compared to estimates that do not account for mitigation and adaptation needs within these sectors.

### Methodology overview for health (SDG 3)
- Empirical approach: a benchmarking approach analogous to Gaspar et al. (2019) to estimate expenditure needed to adapt health systems to climate impacts.
- Three main components of the approach:
  1. Identify countries’ current and future exposure to climate change (Section 1).
  2. Assess countries’ performance in reducing vulnerability given this exposure (Section 2).
  3. Estimate additional cost to achieve “good performance” by benchmarking countries to high performers among peers with similar exposure and projecting needs based on future exposure levels (Section 3).
- The benchmarking approach focuses on adaptation spending in health (no quantified mitigation spending in SDG 3 due to absence of well-established sectoral mitigation standards comparable to education).

### Framing and limitations highlighted
- The paper accounts for both adaptation and mitigation where relevant; for health the focus is on adaptation.
- The study acknowledges inherent limitations and uncertainties:
  - Scientific understanding limits, data quality and availability constraints, model construction choices, and reliance on expert judgment where needed.
  - Uncertainties about availability and costs of future technologies and demographic and socio-economic trajectories.
- Results are intended to inform national policymakers and multilateral stakeholders and to support integration into National Adaptation Plans (NAPs), Nationally Determined Contributions (NDCs), and SDG-aligned budgeting processes.

*Source: IMF Working Paper — "Accounting for climate risks in costing the Sustainable Development Goals", Section I and II (Health background and methodology).*

### 1. Current Health Exposure to Risks Exacerbated by Climate Change

### 1. Current Health Exposure to Risks Exacerbated by Climate Change

### Overview
- Exposure to climate risk: extent to which a country directly or indirectly comes into contact with physical effects of a changing climate, such as rising sea levels, more frequent and intense heat waves, more severe weather events, and changes in precipitation patterns.
- Distinction: exposure (nature and degree of contact with climatic variations) versus vulnerability/readiness (degree to which a system is susceptible or unable to cope).
- Purpose: build an exposure measure that approximates level of exposure using exogenous variables that capture effects of climate change without taking into account response capacity.

### Measurement framework — three principal dimensions
- Dimensions defined by IPCC (2014) and WHO (2015):
  - (i) health impacts of climate-change-exacerbated extreme weather and disaster events;
  - (ii) exposure to climate-sensitive parasitic and viral diseases;
  - (iii) exposure to the risk of malnutrition brought about by climate change.

### a. Health exposure to climate-change-induced extreme weather and disaster events
- Heatwaves:
  - Indicator: change in the number of heatwave exposure events (one exposure event = one heatwave experienced by one person aged over 65 or under 1 year old).
  - Baseline and comparison: change from latest year (2020) measured with respect to cross-year average in reference period (1986–2005).
  - Heatwave definition: a period of at least two days during which both the daily minimum and maximum temperatures are above the 95th percentile of their respective climatologies.
- Wildfire:
  - Uses both model-based wildfire danger and satellite-observed exposure.
  - Climatological wildfire danger: daily ‘very high’ or ‘extremely high’ wildfire danger (a fire danger index score of 5 or 6, respectively, on a range of 1-6) combined with climate and population data.
  - Human exposure measured in person-days (one person-day = one person exposed to a wildfire in one day) using satellite and population data, averaged over a recent four-year average (2017-2020).
- Droughts:
  - Two factors: drought risk and population affected by droughts in recent years (materialized risk).
  - Agricultural drought definition: dry period in which at least 30 percent of the crop area was in stress for more than 10 days.
  - Measured using Agriculture Stress Index (ASI) based on integration of the Vegetation Health Index (VHI); percent pixels in arable areas with VHI value below 35 percent used.
  - Country-year considered in drought if ASI indicates drought in one or more crop seasons.
  - Drought probability: frequency of droughts within the last 30 years.
  - Historical affected population: number of historically affected people per year (absolute and relative) based on last 25 years.
  - Final drought exposure: mean between average annual size of drought-affected population and frequency of drought events.
- Floods, tsunamis, and tropical cyclones:
  - Hazard zones from hazard maps and event return-periods; overlay with population distribution to derive population exposed to hazard type and return-period.
- Population exposure metric:
  - Mean of absolute (unnormalized number of exposed people) and relative (share of total population exposed) population exposure is used after applying a log transformation to the absolute number of exposed people.
  - Each indicator x is rescaled to range from 0 to 10 through a min-max normalization for each country i:
    - 푥̃ 푖 =10−(10 푥 푚푎푥 − 푥 푖 푥 푚푎푥 −푥 푚푖푛)

### b. Exposure to climate-sensitive parasitic and viral diseases
- R0 of Aedes aegypti and Aedes albopictus:
  - Measures environmental suitability for arbovirus transmission (dengue, chikungunya, and Zika).
  - Based on a model capturing influence of temperature and rainfall on vectorial capacity and vector abundance, overlaid with human population density to estimate R0.
  - Variable used: average of indicators for these two vectors.
- Malaria transmission season length:
  - Measured by a threshold-based model incorporating precipitation accumulation, average temperature, and relative humidity to estimate influence on length of transmission season for Plasmodium falciparum.

### c. Exposure to risk of malnutrition attributable to climate change
- Rationale: climate change projected to undermine food security; for wheat, rice, and maize in tropical and temperate regions, climate change without adaptation projected to negatively impact crop production for local temperature increases of 2 degrees celsius.
- Indicator: change in crop growth duration—the time to reach a target sum of accumulated temperatures (ATT)—for maize, winter wheat, spring wheat, rice, and soybean, measured against a 1981-2010 reference period.
  - Crop growth duration measured by the 11-year rolling average of number of days to reach ATT for each crop.
  - Negative change in ATT relative to reference period implies shortened growth duration, more rapid maturing, smaller crops, and lower yield.
- Composite weighted-average indicator:
  - Weights w for crop c reflect importance of each of the n=5 crops in a country, based on value of production Y:
    - 푤 푐 = 푌 푐 ∑ 푌 푐 푛 푐=1
  - Outliers removed: observations x for country i that fulfill either condition:
    - 푥 푖 < 푄1 − 1.5∙ 퐼푄푅
    - 푥 푖 > 푄3 + 1.5∙ 퐼푄푅
    - where Q1 and Q3 are first and third quartiles, and IQR is interquartile range.
  - Min-max normalization applied to non-outlier observations.

### d. Construction of an overall climate-change exposure measure by country, and of exposure groups
- Data sources:
  - Indices on exposure to tsunamis, floods, tropical cyclones, drought, and wildfire from European Commission’s INFORM Risk database.
  - Exposure to heatwaves and indicators of climate suitability for vector-borne transmissions and exposure to malnutrition from the Lancet Countdown.
- Overall exposure measure:
  - Constructed across m=8 exposure types: health exposure to heatwaves, wildfires, cyclones, tsunamis, floods, droughts, malaria, arboviral diseases, food scarcity.
  - Weighted average of indicators for each type; weights w represent gravity of each exposure type e by country, measured as proportion of deaths D attributable to that exposure type:
    - 푤 푒 = 퐷 푒 ∑ 퐷 푒 푚 푒=1
  - Higher final value reflects higher degree of exposure.
- Exposure groups:
  - Countries classified into five groups with an equal number of countries in each: very low, low, medium, high, and very high.

### e. Future projected health exposure to climate change
- Approach: substitute certain variables in existing exposure measure with ND-GAIN indicators under Representative Concentration Pathway (RCP) 4.5 (“moderate emissions scenario”).
- Health exposure to climate-induced extremes:
  - Heatwaves: absolute change in Warm Spell Duration Index (WSDI) from baseline (1960-1990) to future projection (2040-2070). WSDI counts number of days where daily maximum near-surface temperature exceeds the 90th percentile threshold for six or more consecutive days.
  - Floods: percentage change in flood hazard from baseline (1960-1990) to future projection (2040-2070) based on predicted monthly maximum precipitation in five consecutive days.
  - Proxies: current exposure measures used for tsunamis, tropical cyclones, droughts, and wildfire when future projection data unavailable.
- Exposure to climate-sensitive diseases:
  - Projected change in vector-borne diseases indicator: absolute change in malaria Length Transmission Season (LTS) from baseline (1980-2010) to a 2050 projection.
- Exposure to risk of climate-induced malnutrition:
  - Forecasts built using results from five crop models for rice, wheat, and maize; projected change computed as percentage change from baseline (1980-2009) to future projection (2040 to 2069).
- Projected overall exposure metric:
  - Normalize each indicator to 0–10, compute weighted average as in Section 1d.
  - Classify countries into five groups with equal number of countries in each.

### 2. Performance in Reducing the Vulnerability of Health Systems to Climate Risks
- Purpose: construct measures of countries’ performance in reducing vulnerability to climate risks to health. Parallel indicators reflect countries’ use of resources to reduce populations’ vulnerability.
- Four performance indices (normalized using methods analogous to exposure construction, considering both relative and absolute affected populations):
  - Heat related mortality:
    - Captures heat-related mortality in populations older than 65 years.
    - Applies exposure-response function and optimum temperature described by Honda et al. (2014) to daily maximum temperature exposure of population older than 65 to estimate attributable fraction and deaths attributable to heat exposure.
  - Lethality of natural hazards (wildfires, floods, landslides, droughts, and storms):
    - Sums total deaths per country arising from each disaster.
    - Event considered a disaster if at least one criterion met:
      - (i) 10 or more people die from the event,
      - (ii) 100 or more people are affected by the event,
      - (iii) the event triggers a declaration of a state of emergency, or
      - (iv) the event triggers a call for international assistance.
  - Climate-sensitive parasitic or viral diseases:
    - Indicator is sum of deaths due to dengue and lymphatic filariasis, and total malaria cases.
  - Malnutrition attributable to climate change:
    - Measure used is child stunting: number of under-five-year-olds falling below -2 standard deviations (moderate and severe stunting defined as below -2 and -3, respectively) from median height-for-age of reference population.
- Composite performance variable:
  - Weighted-average using weights w as share of deaths from each of n event types p (extreme heat, natural hazards, parasitic or viral diseases, malnutrition):
    - 푤 푝 = 퐷 푝 ∑ 퐷 푝 푛 푝=1
- High performer definition:
  - Country considered a high performer in reducing vulnerability if normalized score > 9.
  - If more than ten countries in an exposure group meet this condition, only the ten highest performers in that group are considered.

*Source: wpiea2024049-print-pdf - 1. Current Health Exposure to Risks Exacerbated by Climate Change*

### 3. Additional Expenditure to Adapt Health Systems to Climate Risks for the Achievement of SDG 3

### 3. Additional Expenditure to Adapt Health Systems to Climate Risks for the Achievement of SDG 3

### Health adaptation spending: tracking and benchmarking methodology
- The Lancet Countdown tracks spending on health adaptation to climate change by country (Romanello et al., 2022). This includes both public and private spending on measures such as health infrastructure development, disease surveillance, disease control systems, and health workforce training, and includes key adaptation measures identified by the IPCC.
- Data acquisition and analysis use a ‘profiling’ system originally developed at Harvard Business School, populating a new taxonomy from the bottom up and including only economic activities where sufficient evidence is available.
- Benchmarking approach to estimate additional climate adaptation spending needs for a country i:
  - Country i’s peer group is the projected exposure group to which it belongs (countries whose health systems have a similar level of future exposure as country i).
  - Country i is benchmarked against the set of ‘good performers’—those most effective in reducing their population’s health systems’ exposure to climate risks given current exposure and vulnerability.
  - i’s additional expenditure to adapt health systems to climate change = difference between the mean of the well-performing peers’ and the country’s per-capita health adaptation costs.
  - Results are reported with these costs expressed as a percent of 2030 GDP.

### Key methodological elements preserved from the source
- Scope: public and private spending on adaptation measures for health systems as defined above.
- Benchmarking uses projected exposure-group membership and current exposure and vulnerability measures.
- Output metric: additional per-capita health adaptation costs expressed as a percent of 2030 GDP.

*Italic source attribution: IMF Working Papers — chapter 3 from wpiea2024049-print-pdf*

---

### III. Education (SDG 4)

### Background
- Climate change poses risks to children’s education and can undo gains toward SDG 4.
- UNICEF calls for “urgent actions to ‘climate-proof’ the education sector and to accelerate climate-resilient and climate-smart education investments and actions” (UNICEF, 2019).
- Direct impacts on education from climate change include:
  - damage of education infrastructure,
  - loss of education material,
  - injury/mortality of students and teachers,
  - impacts on learning from psychosocial stress due to extreme weather events.
- Indirect impacts on educational outcomes include climate effects on food security, livelihoods, air pollution, access to water, health, and energy (UNICEF, 2019, 2022).
- Article 6 of the UNFCCC and Article 12 of the Paris Agreement emphasize education, training, and public awareness on climate change.
- Global education system scale and asset values:
  - 6.6 million schools,
  - 1.6 billion students,
  - 83 million teachers,
  - 41 million administrative staff,
  - asset value of school infrastructure and contents estimated to be US$13.6 trillion (World Bank, 2023).
- Regional and income-group vulnerability:
  - East Asia & Pacific holds the highest total school-asset value; Sub-Saharan Africa has the lowest.
  - Latin America & Caribbean and East Asia & Pacific regions have experienced the most damage from major tropical cyclone events in the past 50 years.
  - Historical records suggest damage in the education sector could surpass 40 percent of the total direct damage of such cyclones (World Bank, 2023).
  - For wind impacts from 20 tropical cyclone events (1998–2018), the average annual loss as percentage of total exposed value of school capital stock is highest for Emerging and Developing Asia region; by income group, EMEs have the highest loss as a share of exposed assets.
- Policy integration gap: Out of the 196 Parties to the UNFCCC, less than one-third mention education in their Nationally Determined Contributions (NDCs) (UNICEF, 2019). Where included, references are often limited to curriculum and public awareness, not addressing multiple pathways through which climate risks undermine educational performance.

### Methodology for additional adaptation and mitigation costs in education
- Empirical basis: UNESCO review found buildings and equipment historically represent 20 to 25 percent of sector spending, second to teachers’ salaries (Beynon, 1997).
- Objective: estimate additional cost of (i) retrofitting/reconstructing current school building stock to improve resilience to major disasters by 2030, and (ii) constructing new buildings following the green schools standard.
- Two components modeled:
  1. Costs of retrofitting existing school buildings to climate-resilience.
  2. Cost of upgrading new buildings to meet the green schools standard.

### 1. The Costs of Retrofitting Existing School Buildings to Climate-Resilience
- Base formula for retrofit/reconstruction cost:
  - RC_i = u ꞏ K_i
    - where u is the average unit (per-square-meter) cost of retrofit measures and replacement/reconstruction options;
    - K_i is the total school building area (in m^2) in country i.
- Derivation of K_i:
  - K_i = N_i ꞏ S_i ꞏ E_i
    - N_i = country-specific norm for required school building area per enrolled student (averaged across school types), estimated from UNESCO reports and other sources; income-region averages applied where data absent.
    - S_i = total number of enrolled students (from World Bank EdStats database).
    - E_i = country-level exposure index ranging from zero to one, with higher values denoting higher levels of exposure to flood hazard (proxies share of school building area exposed to climate risk).
- Unit cost data and proxy:
  - The weighted average unit cost of various retrofit interventions is estimated to be US$156. (Derived from GFDRR Safe Schools report; due to scarcity of granular data, Tonga was used as a proxy for global unit costs of retrofits and reconstruction/replacement measures.)
- Annualized additional spending to make schools climate resilient by 2030:
  - 푅퐴퐶
    푖푇
    = 
    푅퐶
    푖
    ∙푟
    1−
    (
    1+푟
    )
    −푛
  - Parameters specified:
    - r = discount rate assumed to be 5 percent (following World Bank, 2022).
    - n = number of years to the SDG end-year = 10 (given much data pertain to 2020).

### 2. The Cost of Upgrading New Buildings Following the Green Schools Standard
- Rationale: conventional schools designed to local building codes often yield higher operational costs and lower resilience; green schools offer mitigation co-benefits and operational savings.
- Reported co-benefits and operational savings:
  - Typical reductions of around 30 to 50 percent reduction of energy and water use.
- Green cost premium evidence and assumptions:
  - Global review of 17 empirical studies found green cost premium varies between −0.4 to 21 percent (Dwaikat and Ali, 2016).
  - Kats (2010) found a range from 0 to 18 percent; more than three-quarters of analyzed green buildings fell within 0 to 4 percent, with a median value of around 2 percent.
  - Studies of green schools specifically found they tend to cost around 2 percent more on average than conventional schools in the U.S. (Kats, 2006).
  - Acknowledged limitations: green cost premium likely sensitive to local conditions; may be higher in lower-income countries.
- Modeling choice for this study:
  - Apply a uniform 2 percent cost premium for green schools for new investments in school buildings between 2020 and 2030.
  - Formula for additional green premium cost:
    - NC_iT =  E_i K_iST
      - where K_iST = new capital investment spending in schools at time T = 2030 for meeting SDG 4 in country i,
      -  = 0.02 is the green schools cost premium.
- Note on sensitivity: the 2 percent uniform premium is applied while acknowledging actual premiums may vary and estimates may be a lower bound for EMEs and LIDCs.

### Aggregation of education adaptation costs
- Additional total costs in 2030 for country i to meet SDG 4 with climate-resilient educational facilities:
  - TC_iT = RAC_iT + NC_iT

---

### IV. Water and Sanitation (SDG 6)

### Background: climate as the primary channel for water-related impacts
- Water and SDG 6 (clean water and sanitation) are at the frontlines of climate change; the water cycle transmits climate impacts via precipitation, storm surges, floods, droughts, hurricanes, rising seas, and groundwater recharges (Douville et al., 2022; World Bank, 2016).
- The IPCC Sixth Assessment Report confirms significant changes to the world’s water cycle are already underway and will likely grow (IPCC, 2021).
- Existing cost estimates for universal WASH coverage (SDG targets 6.1 and 6.2) exist but have not accounted for climate change impacts (Hutton and Varughese, 2016; Gasper et al., 2019; SDSN, 2019; WRI 2020; Carapella et al., 2023).
- Climate change implications for WASH costs and infrastructure:
  - Water scarcity is predicted to proliferate and worsen in many regions under business-as-usual scenarios.
  - Greater frequency and intensity of climate hazards threaten the capacity of existing infrastructure to deliver WASH services.
- Modeling challenges highlighted:
  - Uncertainty in long-term climate projections complicates estimating additional climate-related costs.
  - Global circulation models (GCMs) were not designed to project changes in the hydrological cycle; hydrology is treated as one element of a larger climate system (Douville et al., 2022).
  - Imprecision increases when models are downscaled to finer spatial scales needed for policy; predicted climate impacts on water availability can be highly uneven across regions (World Bank, 2016).

*Italic source attribution: IMF Working Papers — chapter 3 from wpiea2024049-print-pdf*

### 2. Distribution of Water Stress: Baseline versus Future Trends

### 2. Distribution of Water Stress: Baseline versus Future Trends

### Baseline distribution and near-term projections
- Water stress is defined as the ratio between total water withdrawals and available renewable surface water (Luck et al., 2015).
- World Resources Institute (WRI) modeled potential changes in future demand and supply at global scale for the period 2020 (taken as baseline) to 2050 under two climate scenarios:
  - RCP 4.5 — “moderate emission scenario”
  - RCP 8.5 — “very high emission scenario”
- Baseline (2020) “high” and “extremely high” water stress regions identified include:
  - Mongolia, northern China, parts of south Asia and west Asia, Middle East and Mediterranean region, northern Africa, Sahel, southern Africa, Chile, south-western region of North America, south-western Australia.
- Projected change in water stress from baseline (2020) to future periods (2030–2050) under RCP8.5 / SSP2 shows a large increase in water stress in many regions already highly stressed at baseline, including:
  - much of the Mediterranean, Middle East and North Africa, parts of South Africa, the Murray-Darling River basin in Australia, and the southwestern region of North America.

### Seasonal variability (SV) and implications
- Seasonal variability (SV) is measured as the within-year coefficient of variance between monthly total renewable surface water supplies.
- Increasing SV may mean wetter wet-months and drier dry-months, increasing likelihood of seasonal droughts where reservoir capacity is inadequate.
- Projected increases in SV are notable even in areas not expected to get drier, especially:
  - Sub-Saharan Africa, South Asia, Central Asia, and the southwestern parts of Latin America.
- Policy implication: higher SV, compounded with uncertainty, requires planning for a wide range of possible outcomes.

### Adaptation options and trade-offs
- Supply-side options:
  - Investments in storage infrastructure (reservoirs and dams) smooth variability and provide hydropower and flood protection benefits, but raise ecological and social concerns.
  - Desalination and groundwater pumping expand supply but may conflict with mitigation due to high energy use.
  - Nature-based solutions (wetland preservation, rainwater harvesting, groundwater recharge) are gaining popularity but are highly sensitive to local bio-physical context.
- Demand-side interventions:
  - Behavioral and policy measures to manage demand are important and need integration with supply-side interventions for lasting impact.
- Strategic approach:
  - Decision makers should adopt adaptable and flexible approaches that can respond to new information and changing circumstances, consistent with IPCC WG2 perspectives on impacts and adaptation.

### Methodology overview for costing adaptation (WASH focus)
- Two complementary adaptation cost components estimated:
  1. Cost of adapting to increasing water stress through construction of additional facilities (storage, reuse systems) to provide raw water for industrial and municipal demand.
  2. Cost of making WASH infrastructure more resilient to climate hazards (structural improvements and quality control).
- The costing builds on the Economics of Adaptation to Climate Change (EACC) approach (Margulis et al., 2010), which:
  - Establishes a future socio-economic baseline using consistent population and GDP projections (development trajectory without climate change).
  - Estimates costs accounting for climate risks; additional adaptation costs are the difference.
- Time horizons for demand and climate change projections used: 2030 and 2050.

### Box: Modeling steps in estimating additional water needs to adapt to climate change
- Water demand and supply projections:
  - Climate and Runoff Model (CLIRUN-II) on a monthly time-step.
  - Results aggregated at level of 281 food production units (FPUs) of IFPRI’s IMPACT model; FPUs then aggregated to World Bank regional level.
- Climate scenarios / models:
  - Two GCMs used to capture a wide range: CSIRO_MK3 (extreme dry) and NCAR_CCSM3 (extreme wet), drawn from 26 global climate models that provided projections based on the IPCC A2 emissions scenario from SRES for AR4 (2007).
- Sustainable expansion of supply assumptions:
  - Increased water demand is assumed to be met primarily through reservoir yield by increasing surface reservoir storage capacity, except:
    - when increasing supply from reservoir yield would increase withdrawals to more than 80 percent of river runoff.
    - when the unit cost of supplying water from reservoir yield would be higher than US$0.30/m3. In those cases, supply is met by backstop measures (recycling, rainwater harvesting, or desalination) at an average cost of US$0.30/m3.
  - Ward et al. (2010) global average cost of reservoir storage capacity expansion estimated at US$0.13/m3 (in 2005 USD).
  - Alternative adaptation option cost ranges in Ward et al. (2010):
    - rainwater harvesting: around US$0.03/m3 to US$0.25/m3
    - desalination: US$0.60/m3 for brackish water to US$1.00/m3 for seawater
- Aggregation approaches:
  - Net approach: calculate country-level costs (positive and negative), average for each five-year period, then aggregate to regions.
  - Gross approach: set negative costs to zero before aggregation.
- For this study, the gross cost estimates for NCAR and CSIRO scenarios are used to capture range of possibilities.

### Application to WASH costing
- To extract WASH (household/municipal) component from EACC water sector costs:
  - Use World Bank’s World Development Indicators and WHO/UNICEF JMP for Water Supply and Sanitation to derive share of municipal water use in combined municipal-and-industrial water use for each country.
  - Apply that municipal share to the EACC cost estimate to isolate WASH adaptation costs.

### Existing WASH infrastructure resilience improvement costs
- A high-level project (Hallegatte et al., 2019a, b; Miyamoto, 2019a) assessed engineering solutions and costs for reducing damage probability from climate hazards.
- Key WASH components considered: reservoirs (open and storage tanks), water treatment plants (potable water and wastewater), distribution pipes, sewage network emissaries, water conveyance systems (canals), drainage systems.
- The methodology:
  - Normalize costs of reducing damage probabilities under two major hazards: floods and wind damage.
  - Weight normalized improvement cost for each component by its share in overall water and sanitation system costs (cost shares taken from EPA (2023) and City of Corpus Christi (2018) for water systems, and Jardim et al. (2012) for sanitation systems).
  - Use the weighted sum as the infrastructure resilience cost markup.

### Cross-cutting numeric points and thresholds (preserved)
- Baseline year: 2020.
- Projection periods: 2030–2050.
- Alternative baseline period referenced in Figure 4: 1950-2010.
- FPUs aggregated: 281 FPUs.
- Reservoir withdrawal ecological threshold used: 80 percent of river runoff.
- Backstop average cost when reservoir supply exceeds threshold: US$0.30/m3.
- Global average cost of reservoir expansion (Ward et al., 2010): US$0.13/m3 (in 2005 USD).
- Alternative adaptation option cost ranges:
  - rainwater harvesting: around US$0.03/m3 to US$0.25/m3
  - desalination: US$0.60/m3 to US$1.00/m3
- Climate models used: CSIRO_MK3 and NCAR_CCSM3 (drawn from IPCC A2 SRES for AR4, 2007).
- Aggregation approaches: net approach and gross approach (gross used in this study).

*Source: wpiea2024049-print-pdf - 2. Distribution of Water Stress: Baseline versus Future Trends.*

### 1. Additional Electricity Spending to Adhere to Mitigation Goals

### 1. Additional Electricity Spending to Adhere to Mitigation Goals

### a. Costs arising from a change in the energy mix
- Objective: Derive the 2030 energy mix consistent with containing global warming to 2°C above pre-industrial global average temperatures by mid-century; global greenhouse gas emissions must be reduced by 25 to 50 percent below current levels by 2030 to be on track.
- Data source for projections and current levels: Climate Policy Assessment Tool (CPAT), with country-level projections of electricity generation in 2030 consistent with the ICPF scenario and disaggregation by nine energy sources (coal, oil, natural gas, nuclear, biomass, wind, solar, hydroelectricity, and a residual renewable category).
- Definitions and accounting:
  - G_xit = amount of electricity generated with energy source x in country i at time t = {t0, T}.
  - G_xit_sh = G_xit / sum_x G_xit (share of total electricity generated from source x).
  - con_xit = G_xit_sh ⋅ con_it, where con_it are country-level electricity consumption for time t derived from the non-climate (core) estimation.
  - Counterfactual (business-as-usual, bau) 2030 consumption by source: con_xiT_bau = G_x,i,t0_sh ⋅ con_iT (i.e., if current mix retained).
- Additional mitigation-related spending for country i (cost_i^M):
  - cost_i^M = sum_x [ (con_xiT − con_xiT_bau) ⋅ UC_xiT ]
  - UC_xiT = country- and energy-source-specific investment unit costs in 2030 in US$/kW as estimated by CPAT, including generation, transmission, and distribution costs.
- Transmission and distribution add-on applied in CPAT: 0.04 US$/kWh, or 350.4 US$/kW, is added to each investment unit cost.
- Assumption: apply current and 2030 energy-source shares to the current and 2030 levels of electricity consumption derived in the non-climate (core) SDG costing, to produce comparable estimates of additional spending driven by the shift in the energy mix.

### b. Electricity storage for the additional renewable energy
- Rationale: Increased VRE (wind and solar) raises the need for storage to manage intermittency and satisfy demand during inactive periods.
- Storage estimation approach:
  - Use IRENA (2020) Transformative Energy Scenario to derive a global storage-to-capacity ratio, SCR_t, for current period and for 2030.
  - Storage need: S_it = con_it_vre ⋅ SCR_t (multiply SCR by VRE consumption).
- IRENA 2030 numeric inputs and derived ratio:
  - VRE global capacity in 2030: 5,753 GW.
  - Storage consistent with this increase in capacity in 2030: 745 GWh.
  - SCR_2030 = 745 / 5,753 = 0.129.
- Costing storage:
  - Additional storage cost for country i: cost_i^S = (S_iT − S_i,t0) ⋅ LCOS.
  - LCOS measured in US$/kWh represents the present value of total lifetime costs of storage technology (operation and maintenance, cost of debt, capital spending).
  - LCOS used is the average of PNNL’s estimated unit costs for four battery technologies: lead acid, vanadium, lithium ferrophosphate (LFP), and nickel manganese cobalt (NMC).
  - Projection: account for decreasing cost of battery storage in 2030 by applying Cole and Karkamar (2023) projected storage battery cost for 2030 to the PNNL LCOS estimates.

### c. Electricity adaptation: making networks climate-resilient
- Rationale: Adverse weather events exacerbated by climate risks will increasingly impair electricity networks; adaptation (e.g., flood protection walls, materials with greater fatigue life) increases costs.
- Adaptation cost approach:
  - Apply energy- and climate-event-specific mark-ups from Miyamoto (2019a) as percentage increases α_x to current generation investment costs from CPAT to obtain climate-resilient unit costs:
    - UC_xiT^A = UC_xiT ⋅ (1 + α_x)
  - Total adaptation costs for country i:
    - cost_i^A = sum_x con_xiT ⋅ (UC_xiT^A − UC_xiT) = sum_x con_xiT ⋅ UC_xiT ⋅ α_x
  - Assumption: entire existing and new electricity network needs to be adapted (this makes the estimate an upper bound).
- Total climate-adjusted additional electricity cost for country i:
  - cost_i^CL = cost_i^M + cost_i^S + cost_i^A
- Integration: cost_i^CL is added to 2022 estimates of expenditures to align electricity consumption with economic growth and universal access (the core SDG electricity estimates) to obtain the annual additional spending in percent of GDP to reach SDG 7 when accounting for climate mitigation and adaptation needs.

### Key methodological notes and assumptions (electricity)
- CPAT contains nine energy sources (four non-renewables and four major renewables plus a residual renewables category).
- The analysis produces consumption-by-source estimates by applying shares to consumption totals derived from a non-climate core SDG costing to ensure comparability.
- The LCOS incorporates projected cost declines for battery storage by applying Cole and Karkamar (2023) projections to PNNL estimates.
- Adaptation mark-ups α_x are taken from Miyamoto (2019a); Table 3 (mark-ups for severe floods and winds) is referenced but specific numeric mark-ups are applied only as available in Miyamoto (2019a).

### d. Summary of key numeric inputs cited
- IRENA 2030 VRE capacity: 5,753 GW.
- IRENA 2030 storage: 745 GWh.
- SCR_2030 = 0.129.
- Transmission and distribution add-on in CPAT: 0.04 US$/kWh, or 350.4 US$/kW.
- LCOS basis: average of PNNL’s unit costs for lead acid, vanadium, LFP, NMC; adjusted by Cole and Karkamar (2023) 2030 projection.

---

### VI. Road infrastructure (SDG 9) — key points relevant to climate-adjusted costing

### A. Background and hazards affecting road infrastructure
- SDG 9.1.1 focuses on road access for rural populations; roads are multi-decadal assets designed to historical climatic conditions, making them vulnerable under changing climate.
- Major climate hazards affecting roads: floods (fluvial/river, pluvial/surface, coastal), landslides (mountainous regions), and tropical cyclones (winds and extreme rainfall).
- Tropical cyclones: defined as low-pressure systems forming over tropical waters (25°S to 25°N); winds exceeding 74 mph designated as hurricanes or typhoons. Evidence indicates an increase in the proportion of severe TCs (category 3–5) and projections show further increases (Knutson et al., 2021).

### B. Exposure patterns
- High-income countries account for almost half of total global transport infrastructure length; upper middle-income countries account for almost a quarter.
- Low-income countries’ road infrastructure is particularly exposed to river flooding.
- Road systems in high- and upper-middle-income countries face greater exposure to tropical cyclones due to hazard geography (Caribbean, US Gulf and east coast, eastern China, South Asia, Japan).
- For river and coastal flooding, vulnerability is reduced in some higher-income contexts due to higher flood protection standards.
- Return-periods and design standards:
  - Return-period concept: e.g., 100-year return-period = 1 percent annual exceedance probability (AEP).
  - Norms have shifted in some contexts (US example: from 1 percent AEP to “500-year floods” or 0.2 percent AEP in FEMA guidance).
  - A study found design return-periods of nearly 88 to 95 percent of global transportation assets will become shorter in 2050 relative to 1971–2000, with an average decrease of 25 percent under the mean of RCP4.5 and RCP8.5.

### C. Methodology for road resilience costing
- Exposure data: Koks et al. (2019) subnational estimates of road-type lengths (primary, secondary, tertiary, other) exposed to hazards (fluvial, pluvial, coastal floods, cyclones) by intensity bands and return-periods (10, 20, 50, 100 years).
- Notation:
  - L_ipj_rz = length of road-type r in country i exposed to hazard z occurring with probability p and intensity j.
  - L_ir = sum_z (max_{j,p} L_ipj_rz) — maximum length of road-type r exposed to different intensities and probabilities across hazards, summed across hazards.
- Resilience mark-ups and unit costs:
  - Hallegatte et al. (2019a, b) and Miyamoto (2019a) provide engineering solutions and associated percentage cost mark-ups to reduce damage probabilities.
  - Example: providing barriers and increasing drainage size can reduce flood damage probability of secondary and tertiary roads by half at an additional 3 percent of replacement cost.
  - For primary and other roads where detailed mark-ups are not provided, Koks et al. (2019) estimate a 2 percent markup of replacement costs.
  - For new road investments, apply an average markup of 2.5% from Hallegatte et al. (2019a) to make future assets climate resilient (approximation used due to lack of spatial exposure detail for new assets).
- Application: apply these mark-ups to unit road construction costs used in the core SDG road costing (Carapella, Mogues, Pico-Mejía and Soto, 2023) to estimate adaptation costs for existing and new roads.

### D. Methodological limitations and assumptions (roads)
- Landslides and some other hazards excluded due to limited global data.
- Higher return-periods in the dataset (200, 500, 1,000 years) were not considered because road design standards in most countries are based on 50-year or lower return-periods.
- The 2.5% markup for new roads is an approximation reflecting the economic case for upfront resilience investment as more cost-effective than retrofit.

---

### VII. Results — Global climate-adjusted additional spending for selected SDGs (summary)
- Core (non-climate) updated SDG additional annualized spending needs in 2030: 3.41 percentage points of GDP, equivalent to US$3.02 trillion.
- Climate-related addition across the five SDG areas analyzed: annualized 0.41 percent of global 2030 GDP, or US$358 billion.
- Total additional climate-adjusted SDG spending needs in 2030: 3.82 percent of GDP.
- Interpretation: The climate-related addition (0.41 percent of GDP, US$358 billion) is smaller than the core development spending needs but materially augments SDG resource requirements; a larger share of climate-related expenditure is directed to infrastructure-intensive sectors.

*Source: IMF Working Paper — 1. Additional Electricity Spending to Adhere to Mitigation Goals (excerpt).*

### 1. Additional Climate-related Costs to Meet Selected SDGs, relative to GDP

### 1. Additional Climate-related Costs to Meet Selected SDGs, relative to GDP

### By income group and region
- LIDCs experience, relative to their GDP, the largest augmentation to their SDG costs:
  - Core needs: 16.12 percent of 2030 GDP.
  - Climate-related rise: 1.04 percent of GDP.
  - Total: 17.17 percent of GDP.
- EMEs:
  - Climate-related rise: 0.54 percent of GDP.
  - Total: 5.29 percent of GDP.
- AEs:
  - Additional climate-related SDG spending: 0.15 percent of their GDP.
  - Additional core spending: 0.17 percent of AEs’ GDP.
- Regional highlight:
  - Sub-Saharan Africa (SSA) full climate-adjusted additional resource needs: 20.54 percent of GDP — 1.15 percent of GDP higher than the core expenditure needed ignoring climate risks.
- Note on rounding: "The core and climate-related figures are reported here rounded to the second digit after the decimal, hence may not add up to the total shown here."

### Sectoral breakdown
- Climate-related SDG spending is concentrated in infrastructure sectors, amounting to the majority in each income group.
- The higher the income group, the larger the weight of climate-related resource needs in the infrastructure sectors of electricity, roads, and water and sanitation.
- For LIDCs, ensuring resilience of the education and health sectors is sizeable.
- For EMEs, additional mitigation and adaptation costs related to the electricity sector loom large.

### In absolute terms and cross-study comparisons
- Aggregate absolute needs:
  - EMEs:
    - Core SDG delivery needs: US$2.35 trillion.
    - Additional to address climate risks: US$267 billion.
  - LIDCs:
    - Climate-adjusted additional SDG costs in 2030: US$644 billion.
    - Core estimate (pre-climate adjustment): US$605 billion.
  - AEs:
    - Additional climate-adjusted SDG expenditure needs: US$112 billion.
    - Of which climate-related: US$52 billion.
- Global totals and shares:
  - World needs to allocate an additional 3.8 percent of GDP by 2030 towards the selected SDG goals, which includes an additional 0.4 percent of global GDP to render the SDG sectors resilient and support global emission reduction objectives.
- Comparisons with other studies (methodology/scope caveats noted):
  - Rozenberg and Fay (2019): additional 4.5 percent of GDP for low- and middle-income countries to achieve infrastructure SDGs while staying on track to meet the 2oC goal — lower than this study’s EME climate-augmented cost of 5.3 percent of GDP, but methodological differences exist.
  - Hallegatte et al. (2019): additional cost of making infrastructure climate-resilient amounts to 3 percent of overall investment needs in low- and middle-income countries.
  - This study: "arrive at an addition of approximately 6 and 10 percent of climate-related costs on top of the core SDG needs, in the case of LIDCs and EMEs, respectively."
  - IMF blog (Gaspar, Mansour and Vellutini, 2023) estimated emerging markets and developing economies need $3 trillion annually through 2030 — differences reflect scope and methodological choices.
  - Bhattacharya et al. (2022) estimate incremental needs of US$3.5 trillion by 2030, similar to this study though with methodological variations noted.
- Energy-sector comparability caveat:
  - Other reports focus on gross investments, while this analysis considers net additional investment (difference to maintaining the current energy mix) and is concerned only with the electricity sector rather than other uses of energy.

### Summary findings and policy implications
- Key findings:
  - Considerable additional spending is required to achieve selected SDGs while factoring in climate risks.
  - Additional global allocation needed by 2030: 3.8 percent of GDP, including 0.4 percent of global GDP for climate resilience and emissions.
  - LIDCs face the most significant climate-related SDG costs relative to GDP and already face the largest additional expenditure requirements to meet core SDGs.
  - EMEs bear the highest costs in absolute terms.
  - Climate-related spending needs in the SDGs of focus should be viewed as investments to avoid greater future private and fiscal costs from climate-related shocks.
- Policy recommendations and financing considerations:
  - Need for a context-specific approach to resource allocation and substantial international cooperation and financial support.
  - Financing options and reform agendas highlighted:
    - IMF (2023a): establishing a carbon price floor, complemented by instruments to correct other market failures, can reduce emissions and generate revenue for adaptation.
    - Governments should pursue ambitious reform agendas, mobilize domestic revenue, and improve spending efficiency to increase national resources to meet SDG spending needs.
    - Increasing official aid to the United Nations target of 0.7 percent of gross national income (GNI) would largely cover the financing gap of the four case countries in a referenced study.
    - The Copenhagen Accord (COP15, 2009) stipulated developed countries would provide scaled-up climate finance to developing countries, reaching $100 billion a year in 2020; the IPCC notes climate financing should be "new and additional."
  - Research gaps and future work:
    - Cost-benefit analysis of mitigation and adaptation measures is beyond this study’s scope; future work could compare climate-related expenditure needs to costs of delaying action beyond 2030.
    - Expand analysis to include other SDGs (including SDG 13) and model interdependencies among SDGs to support policy coherence.
    - Recognize sector-specific budget and KPI constraints that limit accounting for interdependencies in practice.

*Source: Authors’ estimates and analysis as presented in the provided content.*

### References

### wpiea2024049-print-pdf - References

### Climate change science, synthesis reports, and mitigation
- Bates, B.C., Z.W. Kundzewicz, S. Wu, and J.P. Palutikof (2008). “Climate Change and Water.” Technical Paper of the Intergovernmental Panel on Climate Change. Geneva: IPCC Secretariat.
- IPCC (2001). Climate Change 2001: Synthesis Report. A Contribution of Working Groups I, II, and III to the Third Assessment Report of the Intergovernmental Panel on Climate Change. Watson, R.T. and the Core Writing Team (eds.). Cambridge University Press, Cambridge, United Kingdom, and New York, NY, USA.
- IPCC (2014). Climate Change 2014: Synthesis Report. Contribution of Working Groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change. Core Writing Team: R.K. Pachauri and L.A. Meyer (eds.). Geneva, Switzerland.
- IPCC (2021). Climate Change 2021: The Physical Science Basis. Contribution of Working Group I to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Masson-Delmotte, V., P. Zhai, A. Pirani, S.L. Connors, C. Peán, S. Berger, N. Caud, Y. Chen, L. Goldfarb, M.I. Gomis, M. Huang, K. Leitzell, E. Lonnoy, J.B.R. Matthews, T.K. Maycock, T. Waterfield, O. Yelekçi, R. Yu, and B. Zhou (eds.)]. Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, pp. 3−32.
- IPCC, 2022: Climate Change 2022: Mitigation of Climate Change. Contribution of Working Group III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [P.R. Shukla, J. Skea, R. Slade, A. Al Khourdajie, R. van Diemen, D. McCollum, M. Pathak, S. Some, P. Vyas, R. Fradera, M. Belkacemi, A. Hasija, G. Lisboa, S. Luz, J. Malley, (eds.)]. Cambridge University Press, Cambridge, UK and New York, NY, USA.
- IPCC, 2023: Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland, pp. 35-115.
- IEA (2023). Net Zero Roadmap: A Global Pathway to Keep the 1.5oC Goal in Reach. International Energy Agency.
- IRENA (2020). “Global Renewables Outlook: Energy Transformation 2050,” International Renewable Energy Agency, Abu Dhabi.
- IRENA (2022). “Renewables Power Generation Costs in 2021,” International Renewable Energy Agency, Abu Dhabi.
- Knutson, T. R., Chung, M. V., Vecchi, G., Sun, J., Hsieh, T-L. and Smith, A. J. P., 2021: Science Brief Review: Climate change is probably increasing the intensity of tropical cyclones. In: Critical Issues in Climate Change Science, edited by: C. Le Quéré, P. Liss, P. Forster.
- Lamb, William, Thomas Wiedemann, Julia Pongratz, and others (2021). A review of trends and drivers of greenhouse gas emissions by sector from 1990 to 2018. Environmental Research Letters, 16:1-32.
- Liu, K., Wang, Q., Wang, M. et al. Global transportation infrastructure exposure to the change of precipitation in a warmer world. Nat Commun 14, 2541 (2023).
- Luck, M., M. Landis, F. Gassert (2015). “Aqueduct Water Stress Projections: Decadal Projections of Water Supply and Demand Using CMIP5 GCMs.” Technical Note. Washington, D.C.: World Resources Institute.
- Douville H, Allan R.P., Arias P.A., Betts R.A., Caretta MA, Cherchi A, et al. (2022). Water remains a blind spot in climate change policies. PLOS Water 1(12): e0000058.

### Adaptation, resilience, and infrastructure costs
- Hallegatte, S., J. Rentschler, and J. Rozenberg (2019a). “Lifelines. The Resilient Infrastructure Opportunity,” World Bank, Washington, DC.
- Hallegatte, S., J. Rentschler, J. Rosenberg, C. Nicolas, C, Fox (2019b). “Strengthening New Infrastructure Assets. A Cost-Benefit Analysis,” Policy Research Working Paper 8896, World Bank, Washington, DC.
- Margulis, S., Hughes, G., Schneider, R., Pandey, K., Narain, U., & Kemeny, T. (2010). Economics of adaptation to climate change: Synthesis report. Washington, D.C.: World Bank Group.
- Margulis, Sergio; Narain, Urvashi (2010). The costs to developing countries of adapting to climate change: new methods and estimates – the global report of the economics of adaptation to climate change study (English). Washington, D.C.: World Bank Group.
- Miyamoto International (2019a). “Overview of Engineering Options for Increasing Infrastructure Resilience,” Background Paper for Lifeline Report, World Bank, Washington, DC.
- Miyamoto International (2019b). “Increasing Infrastructure Resilience – Technical Annex,” Technical annex to background paper above, World Bank, Washington, DC.
- Koks, E. E., Rozenberg, J., Zorn, C., Tariverdi, M., Vousdoukas, M., Fraser, S. A., Hall, J. W., & Hallegatte, S. (2019). A global multi-hazard risk analysis of road and railway infrastructure assets. Nature Communications, 10(1): 2677.
- Ward, Philip J.; Pauw, Pieter; Brander, Luke M.; Aerts, Jeroen, C.J.H.; Strzepek, Kenneth M. (2010). Costs of adaptation related to industrial and municipal water supply and riverine flood protection (English). Development and climate change discussion paper; no. 6 Washington, D.C.: World Bank Group.
- Liu, K., Wang, Q., Wang, M. et al. Global transportation infrastructure exposure to the change of precipitation in a warmer world. Nat Commun 14, 2541 (2023).
- Caltrans (2018). Caltrans Climate Change Vulnerability Assessments: District 4 Technical Report. January 2018.
- FEMA (2022). Federal Flood Risk Management Standard.
- USGS (2023). Floods and Recurrence Intervals. United State Geological Service.
- Ponce, V (2008). On the return-period to be used for design. San Diego State University.

### Water, sanitation, and wastewater economics
- Bates, B.C., Z.W. Kundzewicz, S. Wu, and J.P. Palutikof (2008). “Climate Change and Water.” Technical Paper of the Intergovernmental Panel on Climate Change. Geneva: IPCC Secretariat.
- Hutton, G. and Varughese, M. (2016). Costs of Meeting the 2030 Sustainable Development Agenda Targets on Drinking Water, Sanitation and Hygiene. Washington, DC: World Bank, Water and Sanitation Program.
- Ward, Philip J.; Pauw, Pieter; Brander, Luke M.; Aerts, Jeroen, C.J.H.; Strzepek, Kenneth M. (2010). Costs of adaptation related to industrial and municipal water supply and riverine flood protection (English). Development and climate change discussion paper; no. 6 Washington, D.C.: World Bank Group.
- WRI (2020). Achieving Abundance: Understanding the Cost of a Sustainable Water Future. Authors: Strong, C., Kuzma, S., Vionnet, S., Reig, P. World Resources Institute, Washington D.C.
- IWA (2022). “Reducing the Greenhouse Gas Emissions of Water and Sanitation Services”, IWA Publishing.
- EPA (2023). Drinking Water Treatment Technology Unit Cost Models.
- Stephenson, D. (1998). Water Supply Management. Springer Netherlands. Imprint: Springer.
- Jardim, A. M., D. Imbroisi, J. M. Nogueira, P. H. Z. da Conceição (2012). Economics of wastewater treatment: cost-effectiveness, social gains and environmental standards. Environmental Economics, Volume 3.
- Luck, M., M. Landis, F. Gassert (2015). “Aqueduct Water Stress Projections: Decadal Projections of Water Supply and Demand Using CMIP5 GCMs.” Technical Note. Washington, D.C.: World Resources Institute.

### Health, public health systems, and climate impacts
- Watts, Nick, W Neil Adger, Paolo Agnolucci, Jason Blackstock, Peter Byass, Wenjia Cai, Sarah Chaytor, Tim Colbourn, Mat Collins, Adam Cooper et al. (2015), “Health and climate change: Policy responses to protect public health” Lancet; 386: 1861–1914.
- Fil­ho, Walter Leal, Ulisses M. Azeiteiro and Fátima Alves (2016). “Climate Change and Health: An Overview of the Issues and Needs,” Ch. 1, In: Filho, Azeiteiro and Alves (eds.), Climate Change and Health: Improving Resilience and Reducing Risks. Cham, Heidelberg, New York, Dordrecht, London: Springer.
- Romanello, Marina, Claudia Di Napoli, Carole Green, Harry Kennard, Pete Lampard, Daniel Scamman, Maria Walawender, et al. (2023) “The 2023 report of the Lancet Countdown on health and climate change: The imperative for a health-centered response in a world facing irreversible harms”, Lancet, 400: 1619-1654.
- Setoguchi, Soko, Desmond Leddin, Geoffrey Metz, and M. Omary (2022). Climate Change, Health, and Health Care Systems: A Global Perspective. Gastroenterology, 162: 1549-1555.
- WHO (2015). “Operational framework for building climate resilient health systems” Geneva: World Health Organization.
- Kruk, M. E., Myers, M., Varpilah, S. T., & Dahn, B. T. (2015). What is a resilient health system? Lessons from Ebola. The Lancet, 385(9980): 1910-1912.
- Wu, Xiaoxu, Yongmei Lu, Sen Zhou, Lifan Chen, Bing Xu (2016). “Impact of climate change on human infectious diseases: Empirical evidence and human adaptation,” Environment International, 86: 14-23.
- Benevolenza, Mia A. and LeaAnne DeRigne (2019) “The impact of climate change and natural disasters on vulnerable populations: A systematic review of literature,” Journal of Human Behavior in the Social Environment, 29(2): 266-281.

### Education infrastructure, green buildings, and schools
- Beynon (1997). Physical facilities for education: what planners need to know. UNESCO, International Institute for Educational Planning.
- Kats, G. (2006). Greening America’s schools: Costs and benefits. A capital E report.
- Kats, G. (2010). Greening Our Built World: Costs, Benefits, and Strategies, Island Press, Washington, DC, USA, 2010.
- Kats, Gregory (2005). National Review of Green Schools: Costs, Benefits, and Implications for Massachusetts. A Report for the Massachusetts Technology Collaborative.
- UNICEF (2011). “Strategic roadmap for developing Green Schools Program.” United Nations Children’s Fund.
- UNICEF (2019). “It is getting hot: Call for education systems to respond to the climate crisis. Perspectives from East Asia and the Pacific.” United Nations Children’s Fund.
- Department of Education (DOE) (2014). Area guidelines for mainstream schools. Building Bulletin 103, June. United Kingdom.
- Vickery, D. (1985). Norms and standards of educational facilities, Division of Educational Policy and Planning. UNESCO.
- Nge U Nyi Hla, Daw Win Maw, and U Tet Tun (1992). Norms and Standards of Education Facilities. Myanmar Ministry of Education, UNDP and UNESCO. Working paper Series, UNESCO Digital Library.
- World Bank (2023). Global Baseline of School Infrastructure.
- World Bank (2020). Overview Note: GPSS Global Risk Model. Global Program on Safe Schools. World Bank GPURL D-RAS Team.
- World Bank (2022). “Strengthening the Resilience of Public Facilities in Samoa, Tonga, and Vanuatu (SRPF).” Final Report on the Assessment of Schools in Tonga.

### SDG costing, public finance, and IMF/World Bank analyses
- Carapella, Piergiorgio, Tewodaj Mogues, Julieth Pico-Mejía, and Mauricio Soto (2023). “How to Assess the Spending Needs of the Sustainable Development Goals: The 3rd Edition of the IMF SDG Costing Tool,” IMF How-To-Notes Series.
- Tiedemann, Johanna, Veronica Piatkov, Dinar Prihardini, Juan Carlos Benitez, and Aleksandra Zdzienicka (2021). “Meeting the Sustainable Development Goals in Small Developing States with Climate Vulnerabilities: Cost and Financing” IMF Working Paper No. WP/21/62. International Monetary Fund.
- Mogues, Tewodaj, Nick Carroll, Isaura García Valdés, and Mariano Moszoro (2023). “Namibia: Expenditures in 2030 to Support the Sustainable Development Goals”. Technical Report. International Monetary Fund.
- Garcia-Escribano, Mercedes, Pedro Juarros, and Tewodaj Mogues (2022). Patterns and Drivers of Health Spending Efficiency. IMF Working Paper WP/22/48. International Monetary Fund, Washington DC.
- Gaspar, Vitor, David Amaglobeli, Mercedes Garcia-Escribano, Delphine Prady, and Mauricio Soto (2019). “Fiscal Policy and Development: Human, Social, and Physical Investment for the SDGs,” IMF Staff Discussion Note SDN/19/03.
- Gaspar, Vitor, Mario Mansour, and Charles Vellutini (2023). “Countries Can Tap Tax Potential to Finance Development Goals,” IMF Blog. International Monetary Fund, Washington DC.
- Bhattacharya A et al. (2022) “Financing a big investment push in emerging markets and developing economies for sustainable, resilient and inclusive recovery and growth,” London: Grantham Research Institute on Climate Change and the Environment, London School of Economics and Political Science, and Washington, DC: Brookings Institution.
- SDSN (2019). In: Sachs, J, McCord, G, Maennling, N, Smith, T, Fajans-Turner, V, SamLoni, S (Eds.), SDG Costing and Financing for Low-Income Developing Countries. Sustainable Development Solutions Network (SDSN), New York, Paris, and Kuala Lumpur.
- Sachs, J.D., G. Schmidt-Traub, M. Mazzucato, D. Messner, N. Nakicenovic and J. Rockström (2019). Six transformations to achieve the sustainable development goals. Nature Sustainability; 2: 805-814.
- Hallegatte, S., J. Rentschler, and J. Rozenberg (2019a). “Lifelines. The Resilient Infrastructure Opportunity,” World Bank, Washington, DC.
- Hattle, A. and J. Nordbo (2022). “That’s Not New Money: Assessing how much public climate finance has been “new and additional” to support for development”. CARE Denmark.
- IMF (2023a). Fiscal Monitor—Climate Crossroads: Fiscal Policies in a Warming World. October 2023. International Monetary Fund, Washington DC.
- IMF (2023b). Global Financial Stability Report: Financial and Climate Policies for a High-Interest-Rate Era. October 2023. International Monetary Fund, Washington DC.
- Parry, Ian, Simon Black, and James Roaf (2021). “Proposal for an International Carbon Price Floor Among Large Emitters,” Staff Climate Note 2021/001. International Monetary Fund, Washington DC.
- Black, Simon, Jean Chateau, Florence Jaumotte, Ian Parry, Gregor Schwerhoff, Sneha Thube, and Karlygash Zhunussova (2022). “Getting on Track to Net Zero: Accelerating a Global Just Transition in This Decade,” Staff Climate Note 2022/010. International Monetary Fund, Washington DC.
- Daniel, J., A. Banerji, R. Neves, D. Prihardini, C. Sandoz, A. Zdzienicka (IMF), A. Blackman, S. Esler, N. Palu, V. Piatkov, and T. Moeaki (2020). Tonga: Climate Change Policy Assessment. International Monetary Fund.
- Nose, Manabu (2021). “Solomon Islands—Selected Issues: Spending Needs for Achieving SDGs with Climate Resilience,” International Monetary Fund.
- Mogues, Tewodaj, Nick Carroll, Isaura García Valdés, and Mariano Moszoro (2023). “Namibia: Expenditures in 2030 to Support the Sustainable Development Goals”. Technical Report. International Monetary Fund.

### Sectoral studies and methodological resources
- Griggs, D. J., Nilsson, M., Stevance, A., & McCollum, D. (2017). A guide to SDG interactions: from science to implementation. International Council for Science, Paris.
- Nerini, F.F., B. Sovacool, N. Hughes, et al (2019). Connecting climate action with other Sustainable Development Goals. Nature Sustainability; 2: 674-680.
- Swain, R. B. and S. Ranganathan (2021). Modeling interlinkages between sustainable development goals using network analysis. World Development, 138, 105136.
- Laumann, Felix, Julius von Kügelagen, Thiago Uehara, and Mauricio Barahona (2022). “Complex interlinkages, key objectives, and nexuses among the Sustainable Development Goals and climate change: a network analysis” Lancet Planet Health 6(e): 422-430.
- Haile, B., C. Azzarri, and H. E. Ahn (2021). “Literature Review on Linkages between Child Nutrition and Economic Growth,“ IFPRI-MCC Series, Technical Paper 6. International Food Policy Research Institute, Washington DC.
- Johansen, D.F. and R.A. Vestvik (2020). “The cost of saving our ocean – estimating the funding gap of sustainable development goal 14” Marine Policy, 112: 1-8.
- Kapsoli, Javier, Tewodaj Mogues, and Geneviève Verdier (2023). Benchmarking Infrastructure Using Public Investment Efficiency Frontiers. IMF Working Paper WP/23/101. International Monetary Fund, Washington DC.
- Cole, Wesley and Akash Karmakar. (2023). Cost Projections for Utility-Scale Battery Storage: 2023 Update. Golden, CO: National Renewable Energy Laboratory. NREL/TP-6A40-85332
- Caltrans (2018). Caltrans Climate Change Vulnerability Assessments: District 4 Technical Report. January 2018.
- City of Corpus Christi (2018). Water Utilities: Cost of Service and Rates.
- FEMA (2022). Federal Flood Risk Management Standard.
- National Oceanic and Atmospheric Agency (2023). Hurricanes; Frequently asked Questions.
- USGS (2023). Floods and Recurrence Intervals. United State Geological Service.

### Other assessments, reviews, and reports
- Aitsi-Selmi, A., Egawa, S., Sasaki, H., Wannous, C., & Murray, V. (2015). The Sendai framework for disaster risk reduction: Renewing the global commitment to people’s resilience, health, and well-being. International Journal of Disaster Risk Science, 6(2), 164-176.
- Benevolenza, Mia A. and LeaAnne DeRigne (2019) “The impact of climate change and natural disasters on vulnerable populations: A systematic review of literature,” Journal of Human Behavior in the Social Environment, 29(2): 266-281.
- Dwaikat, L. N., & Ali, K. N. (2016). Green buildings cost premium: A review of empirical evidence. Energy and Buildings, 110, 396–403.
- Ponce, V (2008). On the return-period to be used for design. San Diego State University.
- Tharme, R. E. (2003), A global perspective on environmental assessment: Emerging trends in the development and application of environmental flow methodologies for rivers, River Res. Appl., 19, 397–441.
- Stern, Nicholas (2007). The Economics of Climate Change: The Stern Review. Cambridge and New York: Cambridge University Press.
- Wils A. (2015) Reaching education targets in low and lower middle-income countries: Costs and finance gaps to 2030.
- Nunn, P., & Kumar, R. (2020). “Pacific Islands must stop relying on foreign aid to adapt to climate change, because the money won’t last.” The Conversation, 31 July 2020.
- Kim, Jung Eun (2019) Sustainable energy transition in developing countries: the role of energy aid donors, Climate Policy, 19:1, 1-16.
- IWA (2022). “Reducing the Greenhouse Gas Emissions of Water and Sanitation Services”, IWA Publishing.

*Accounting for Climate Risks in Costing the Sustainable Development Goals — Working Paper No. WP/2024/049*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2024/english/wpiea2024049-print-pdf.pdf_