Physical Climate Tail Risks in Global Logistics REITs

As sustainability targets, net-zero commitments and climate change risks move higher on the agenda of investors, companies and other stakeholders and reporting standards are being developed, more focus is being put to understand the potential impact of these transitions and hazards on asset pricing to avoid a significant risk of mispricing of investments and portfolios.

In this note we evaluate physical climate risks in global logistics real estate assets, as the sector is critical to economic activity, global trade, changing consumption patterns and investment allocations and any significant risks would have effects not only on investment exposures to listed real estate companies and portfolios but also across wider supply chains. Our evaluation is based on assets owned and operated by listed real estate companies (developed markets listed companies) excluding development land; nearly 8,800 assets representing c. 200M sqm / 2.1Bn sqft of global logistics capacity. Assets located in North America account for ca 44% of global capacity, followed by Europe at 26%, Asia at 22%, Oceania at 7% and South America at 1%.

Global logistics asset locations owned by listed real estate companies, c. 200M sqm / 2.1Bn sqft capacity

Size of geographic markets by sqm of asset locations

Source: Klimetrics, company information

Across asset portfolios both the number of assets and the distribution of size of individual assets matter as both smaller number and larger individual assets imply higher impact on risk distribution and potential financial consequences of any physical climate risks events. The figure below illustrates that while the typical size of assets is relatively similar across North America and Europe, and somewhat higher across Oceania (+57% as average and +38% as median), assets across Asian markets are significantly larger at c 2.4x the North America/Europe average (2.5x the North America/Europe median), and Asian markets have significantly more “very large assets”.

Typical distribution of asset size by sqm and very large assets by geographic market

Note: Extreme outlier assets by area in each region are excluded in the figure above for scale of illustrating typical asset size distributions within respective region but are included in climate risk calculations.


Methodology

Klimetrics provides distributions of a range of both acute and chronic physical climate risks based on CMIP and other official scientific climate models. All climate hazards are calculated over land measuring both the occurrence of the specific event (projected non-zero event) and its severity. We use RCP8.5 scenario for the analysis also known as ‘business as usual’ which represents the high-emissions scenario (SSP585: with an additional radiative forcing by the year 2100, SSP585 scenario represents the upper boundary of the range of scenarios, it can be understood as an update of the CMIP5 scenario RCP8.5, now combined with socioeconomic reasons).

Climate hazards evaluated are the following targeting time horizons until 2030 and 2050;

  • Sea Level Rise (SLR) and Coastal Floods: extent and inundation depth, 100-year flood return period.

  • Riverine Floods: extent and inundation depth, 100-year flood return period.

  • Extreme Heat: number of days with maximium temperatures above 30°C and more extreme above 35°C, calculated from daily, biased-corrected climate data

  • Extreme Cold: number of days with minimum temperatures below 0°C and more extreme below -5°C, calculated from daily, biased-corrected climate data

  • Heavy Precipitation: number of periods with days of heavy daily precipitation above 20mm, calculated from daily, biased-corrected climate data

  • Fire Weather Index (FWI); number of periods with index values above 50, FWI is a combination of Initial spread index and Build-up index and is a numerical rating of the potential frontal fire intensity; indices are based on meteorological conditions favourable to the start, spread and sustainability of fires.

  • Hurricane Index (HI): Hurricane Index is calculated by aggregating all hurricane paths at a given location accounting for the intensity of each based on the Saffir-Simpson Hurricane Scale; data is provided by IBTrACS by the US NOAA.

To assign risk exposures for individual assets (asset events), we map the specific geolocation of each asset to individual hazards (layers) measuring the projected occurrence of the event at the specific location and the severity of the event,

Secondly, we map all asset events onto the relevant hazard distribution and assign a risk group to each asset based on deciles of that hazard distribution, 10 being the most severe projected exposure.

Results

Sample results for floods are presented here, please contact Klimetrics for detailed risk report, data or climate hazards information.

Based on the climate scenarios and time periods outlined above, the global logistics sector as defined by assets owned and operated by listed real estate companies, is projected to experience some impact related to physical climate change risks, however it is projected be limited.

  • SLR and Coastal Floods are projected to affect c. 1.8% of global logistics capacity by sqm area by 2030; the more severe risk groups of 6 and higher represent c. 1.1% of capacity at risk and the extreme tails of risk groups 9 and 10 account for 0.4%. Extending the risk assessment horizon to 2050, the projected risks are c. 2.0% overall, c. 1.3% for the more severe risk groups and 0.5% for the extreme tails risk groups.

  • Riverine Floods are projected to affect c. 16.3% of global logistics capacity by sqm area by 2030; the more severe risk groups of 6 and higher represent c. 11.4% of capacity at risk and the extreme tails of risk groups 9 and 10 account for 3.7%. Extending the risk assessment horizon to 2050, the projected risks are c. 16.4% overall, c. 11.6% for the more severe risk groups and 4.0% for the extreme tails risk groups.

  • Despite the higher projected impact of riverine floods in terms of potential occurrences affecting a higher number of assets and larger share of global capacity, as illustrated in the figure below, the two flood risks have significantly different severity profiles with riverine floods showing a much lower inundation depth for the vast majority of potential asset events with only a limited number of locations showing significant jump in severity levels. If we segment potential financial risks based on inundation depth, e.g. 1m or higher for the longer 2050 time horizon, SLR and coastal floods impact c. 1.4% of global capacity compared to riverine floods at c. 2.4%.

  • In terms of potential financial value at risk given calculations are based on projections of 100-year return periods (i.e. there is a 1% probability of the frequency of the respective floods being exceeded in any one year), a high-level estimate would be c. 1-3bps of global capacity being at risk of floods in any given year.

  • Finally, even assets that experience a 1m or higher flooding would likely not be expected to cause full economic loss.

SLR and Coastal Floods by risk group and inundation depth

Source: Klimetrics, climate models

Riverine Floods by risk group and inundation depth

Source: Klimetrics, climate models

Get the report or for data inquiries


About Klimetrics

Klimetrics is provided by Kania Advisors, a quantitative analytics and technology firm focused on real assets. Kania Advisors is formally supporting the TCFD framework and is included in the TCFDs list of supporting companies in the Information Technology category.

Previous
Previous

Physical Climate Tail Risks in US CMBS

Next
Next

Correcting Tropical Cyclone Paths for Risk Assessment