- The World Bank forecasts that by 2045, urban areas are expected to house 70% of the global population and account for 80% of economic activity, leaving them highly exposed to social and economic disruptions from extreme heat.
- Interpreting extreme-heat risk profiles is not always an intuitive process, but hazard percentiles compared across climate scenarios can help introduce a quantitative lens.
- Adaptation costs related to extreme heat may increase four-fold for companies if the world stays closer to its current trajectory of warming, compared to a 1.5°C scenario.
Institutional investors, employers and governments may benefit from understanding the risks of business disruptions from extreme heat by comparing outcomes across a range of climate scenarios. For example, if global temperatures increase by 3°C — close to the world’s current trajectory — the hottest day in summer today could become a mainstay of the season for an average of 30 to 50 days a year by the latter part of this century.1 Urban areas are key to understanding extreme-heat risks because they are the most populated and where most economic activity is generated.2
Business disruption as a risk metric
One way to gauge extreme-heat-related risks is through the number of days per year a city experiences temperatures above 30°C (86°F). This metric — extreme heat days — can be used to approximate exposure to chronic risks such as business disruptions related to lower labor productivity, higher cooling costs and increased infrastructure degradation. We used the MSCI Climate Value-at-Risk (Climate VaR) model to see how many additional extreme heat days there could be in 10 major cities by 2050 compared to 2022 under both a cooler scenario (1.5°C temperature rise) and a warmer one (3°C temperature rise).3 Rather than just looking at cities, we also wanted to compare companies with large real-estate holdings that may face similar levels of extreme heat. Thus, we also included the sector average of the MSCI ACWI Index’s4 real-estate constituents in our analysis.5
Additional extreme heat days in 2050 compared to 2022
Data as of April 26, 2023. Source: MSCI ESG Research
In the 3°C scenario, our model showed that the 10 cities would experience an average of 12 more extreme heat days a year in 2050 compared to 2022. For New York City, London, Los Angeles, Seoul, Shanghai, Moscow and Beijing this figure was at least 20% higher than in 2022. At a broader level, the real-estate sector constituents of the MSCI ACWI Index saw an average rise of 18 extreme heat days, a 29% increase over 2022.
This change can also be viewed in percentile terms, comparing future outcomes against current percentiles.6 For example, New York City’s increase in hazard percentile in a 1.5°C scenario already places it where Honolulu was in 2022. As New York City and Tokyo exceed the 50th percentile in a 3°C scenario, this means they had a higher projection of extreme heat days than half of the 9,123 constituents in the MSCI ACWI Investable Market Index (IMI).7 By contrast, London was one of the least affected in terms of shifts in hazard percentiles across our compared scenarios.
As a group, the real-estate constituents of the MSCI ACWI Index increased from the 50th percentile in 2022 to the 55th and 60th percentiles across the 1.5°C and 3°C scenarios, respectively, suggesting their current holdings have an increased exposure to extreme heat over time.
Relative shifts in hazard percentiles for extreme heat in 2050 compared to 2022
Data as of April 26, 2023. Source: MSCI ESG Research
Adaptation costs could quadruple in a 3°C world
One crucial aspect for governments and investors contemplating different warming scenarios is to factor in the magnitude of increase in costs from physical risks materializing, and not just consider energy-transition-related risks in their analysis. The MSCI Climate VaR model estimates that constituents across all industries of the MSCI ACWI Index could face four times more extreme-heat-related costs in a 3°C scenario vs. a 1.5°C scenario.8
The energy, healthcare and infrastructure sectors are expected to bear the brunt of adaptation costs from extreme heat.9 Regions such as Western Europe may have to start building more cooling solutions. An increase in cooling needs is likely to raise peak energy demand during the summers, adding complications to grids that typically prepare for winter peaking. For healthcare, extreme heat could lead to more hospital visits from vulnerable populations such as the elderly, increase occupational-related heat hazards and the viability of habitats for disease vectors. Transport infrastructure could suffer more damage due to asphalt melting, while buildings may need costly retrofits to incorporate proper ventilation and cooling. These infrastructure costs are also likely be compounded by reduced labor productivity from industries such as construction that primarily operate outdoors.
Beating the heat
Preparing for physical climate risks like extreme heat may require investors to look beyond typical textbook factor- or sector-trend explanations, as impacts are more likely to be linked to the geographic locations of assets than attributes of companies, industries or asset classes. Climate metrics informed by a geospatial approach offer an alternative lens to characterize this risk, allowing investors to more comprehensively assess their total portfolio exposure to physical climate risks.
1“Feeling the Heat in the Extremes.” NASA Earth Observatory, Sept. 16, 2022. The study examined the continental U.S. and compared modeling results from 2074-2099 against 1980-2005.
2“Development at a Glance.” World Bank, Oct. 6, 2022.
3We focused our analysis on the 10 cities with the highest Gross Domestic Product in Purchasing Power Parity (GDP/PPP). “Mapped: Distribution of Global GDP by Region.” Anshool Deshkmukh, Visual Capitalist, Oct. 12, 2021.
4135 MSCI ACWI Index real-estate constituents as of April 25, 2023. Sector definitions based on Global Industry Classification Standard (GICS), the global industry classification standard jointly developed by MSCI and Standard & Poor’s.
5Scenarios were based on the Network for Greening the Financial System’s (NGFS) 1.5°C scenario: REMIND NetZero, 3°C scenario: REMIND National Determined Contributions (NDC). An in-depth explanation of their methodology can be found at: “NGFS Climate Scenarios for central banks and supervisors.” NGFS, September 2022.
6For the full methodology of how we compute risks for extreme heat, please refer to “MSCI Climate VaR Methodology Part 4: Physical Climate Risk.” MSCI ESG Research, October 2022, and “Regional Hazard Methodology.” MSCI ESG Research, April 2023, (both client access only).
7Hazard percentiles are first calculated for each location and then aggregated by revenue allocation share to issuer (company) level hazard percentiles. Based on MSCI ACWI IMI Index constituents as of April 25, 2023.
8MSCI’s Climate VaR model estimates that under a 1.5°C temperature rise, constituents of the MSCI ACWI Index could be facing a present value of USD 1 trillion worth of costs. These costs could go up to USD 4.1 trillion in a warmer 3°C scenario.
9Vikki Thompson et al. “The most at-risk regions in the world for high-impact heatwaves.” Nature Communications 14, 2152 (2023).