The hidden transmission channel of physical climate risk: why supply chain matters?

Corporate physical climate risk assessment typically starts with the location of the company’s own assets. Facilities are located and assessed against climate hazard layers, such as floods, windstorms, cyclones, and wildfires, to estimate potential damage at each site under a given climate scenario and time horizon. This bottom-up, asset-level view is an essential foundation because physical climate events can be localised, and facilities owned by the same company can face materially different risk profiles depending on their location and asset characteristics.

This asset-level perspective, however, captures only one part of a company’s exposure. Firms operate within broader economic networks and depend on the continuous flow of raw materials, intermediate goods, energy, transport and logistics services. These inputs often originate from suppliers and infrastructure systems that sit outside the company’s own operational footprint, but still influence its ability to produce, deliver and generate revenue.

A food manufacturer provides a useful illustration. Its production sites may remain fully operational under a given climate scenario, while coffee, dairy or cocoa suppliers may face climate-related disruption from heat stress, drought or changing rainfall patterns. In that case, the company may experience higher reduced availability of key materials or production delays, even though its own facilities have not suffered direct damage from climate events. At portfolio level, this broader view becomes important for differentiating companies with similar physical footprints but different supply chain dependencies. Two firms may appear comparable when assessed only through their owned assets yet face different levels of climate exposure because their sourcing structures, supplier locations and dependence on climate-sensitive inputs differ. A comprehensive physical climate risk framework therefore needs to combine both direct asset-level exposure with the indirect channels through which supply chain disruptions can affect corporate performance.

We model corporate physical risk as the combination of three complementary channels. Each channel is expressed in the same unit, a damage ratio, interpretable as the share of company value impaired at a given horizon under a given scenario, which makes them directly comparable and additive into a single consolidated measure.

2.1: Chronic risk: a structural exposure
Chronic physical risk refers to the gradual and persistent effects of climate change on the environments in which companies operate. These effects include rising average temperatures, changing rainfall patterns, increasing water stress and sea-level rise. Unlike extreme weather events, chronic risks develop over time and can progressively affect economic activity by reducing labour productivity, increasing cooling and operating costs, and placing additional pressure on infrastructure and natural resources.
The framework assesses chronic risk using the chronic damage estimates provided by the Network for Greening the Financial System (NGFS) Climate Scenarios[1]. These estimates quantify how long-term climate change may affect economic activity across different countries under alternative climate pathways and time horizons. They provide a forward-looking view of the potential deterioration in each country’s operating environment and ensure consistency with a widely used climate-scenario framework.
This country-level information is then translated into a company-specific assessment based on the geographic distribution of the company’s facilities. Greater weight is assigned to countries where the company has a larger operational presence. As a result, two companies facing the same global climate scenario may be affected differently because their activities are located in different countries.
This approach recognises that chronic climate risk is broad and structural. A company concentrated in relatively resilient countries may experience a more limited impact, while a company with significant operations in countries exposed to heat, water stress or other long-term climate pressures may face a materially greater effect.

2.2: Acute risk: a company-specific shock
Acute physical risk captures the impact of extreme weather events, such as floods, windstorms and wildfires. These events can cause direct damage to buildings, equipment and infrastructure, and may also lead to business interruption or temporary loss of operating capacity.
The assessment uses a global facility-level dataset covering around 750,000 sites across sectors and regions. Each facility is identified at the latitude-longitude level and expected damages are assessed against forward-looking hazard projections for events such as floods, windstorms, cyclones and wildfires, using Forvis Mazars proprietary models. These assessments rely on climate scenarios known as RCPs, specifically RCP 8.5, RCP 4.5 and RCP 2.6, although availability depends on the underlying hazard model. RCPs represent emissions trajectories and their impact on global warming. RCP 2.6 assumes a rapid decline in emissions with strong mitigation, leading to about 1.5°C of warming by 2100. RCP 4.5 reflects moderate emissions and partial mitigation, resulting in roughly 2–3°C of warming by 2100, while RCP 8.5 corresponds to high emissions with limited mitigation, potentially causing 4°C or more of warming. In this manner, higher RCP values indicate more severe climate impacts[2]. This granular view is essential because acute physical climate risk can vary materially between sites, even within the same company, depending on local conditions and asset characteristics. This allows the framework to estimate potential facility-level damage under different climate futures before aggregating the results to the company level. 

The methodology combines two components. First, hazard information is used to understand the likelihood and severity of climate events at each location. For example, this may include projected flood depths, wind speeds or wildfire conditions. Second, vulnerability information is used to translate the severity of those events into an estimate of potential damage. In simple terms, the framework considers both the kind of event which could occur at a site and how damaging that event would be for the asset located there. 

The results are first calculated at facility level and then aggregated to the company level. This produces a company-specific view of direct physical risk, reflecting the location, hazard exposure and vulnerability of the company’s own assets. 

Relationship Between RCPs and NGFS Scenarios 

The NGFS is a coalition of central banks and financial supervisors established to strengthen the role of the financial system in managing climate-related risks. One of its most widely used contributions is the development of a set of forward-looking climate scenarios that provide a common analytical framework for assessing climate-related financial risks. These scenarios combine assumptions on climate policy, technological developments, energy systems and physical climate impacts to produce consistent projections of macroeconomic and financial variables over long time horizons. They are structured around different climate futures, ranging from orderly transitions with limited warming to "hot house world" pathways characterised by insufficient mitigation and severe physical impacts. As a result, NGFS scenarios have become a reference framework for climate stress testing, risk management and scenario analysis across the financial sector. 

NGFS scenarios are often discussed alongside the RCPs developed by the Intergovernmental Panel on Climate Change (IPCC), but both frameworks serve different purposes. RCPs are climate-science pathways that describe alternative trajectories of greenhouse-gas concentrations and the resulting radiative forcing of the Earth's climate system. They provide the physical climate assumptions used by many hazard models. NGFS scenarios build on broader socio-economic narratives and translate climate outcomes into macroeconomic and financial consequences. In practice, there is no one-to-one correspondence between the two frameworks, although approximate mappings are commonly used. For example, NGFS Net Zero 2050 and Below 2°C scenarios are broadly associated with warming outcomes between the former RCP 2.6 and RCP 4.5 pathways, while Current Policies and other "hot house world" scenarios produce temperature increases closer to those traditionally associated with RCP 6.0 and RCP 8.5. NGFS scenarios can therefore be viewed as an extension of the climate-science framework, linking physical climate outcomes to economic and financial risk assessment. 

2.3: Supply chain risk: a network-driven exposure 

Supply chain risk captures a different source of physical climate exposure: the impact of climate events on the suppliers and production networks that a company depends on. A company may avoid direct damage to its own facilities but still experience disruption if key suppliers are affected or climate-sensitive inputs become harder or more expensive to obtain. 

This channel is particularly important because it cannot be assessed by looking only at the company’s own assets. A map of owned facilities may show limited exposure, but the company may still rely on suppliers located in highly exposed regions or on sectors that are sensitive to climate conditions. In this sense, the company’s climate exposure depends on the wider economic network that supports its operations. 

Our framework assesses this risk by linking physical climate impacts to production networks. The direct climate shocks affecting sectors are first estimated using the same underlying climate assumptions as the direct physical-risk assessment. These shocks are then transmitted through economic relationships between sectors using input-output data that describe how industries buy from and sell to one another. 

The result is a supply chain risk measure that reflects how climate disruption elsewhere in the economy may affect a company through its upstream dependencies. This makes it possible to compare supply chain exposure with the chronic and acute components of physical risk and to assess total climate exposure in a consistent way. 

The key insight is that a company can be operationally exposed to climate risk even when its own facilities are resilient. For many sectors, especially those that depend on climate-sensitive inputs, the supply chain may represent a material part of total physical climate risk. 

The supply chain channel describes how physical climate shocks affecting one part of the economy can translate into impacts for companies elsewhere. The process follows three main steps: climate-related damage affects productive activity in specific sectors and regions; these disruptions are transmitted through economic relationships between sectors; and the resulting exposure is allocated to companies based on their sectoral and geographic footprint. 

In the climate framework the supply chain channel is not treated as a separate assessment, but as an extension of the same climate scenarios, hazard information and damage estimates used in the chronic and acute physical risk analysis. This allows indirect supply chain exposure to be assessed in a way that is comparable with the chronic and acute components described in Section 2. 

3.1: From physical damage to sector shocks 

The first step is to translate physical climate impacts into shocks to productive activity by sector and region. To maintain consistency with the broader framework, the supply chain assessment uses the same climate scenarios, hazard information and damage estimates described in Section 2, rather than introducing a separate set of assumptions. Both chronic and acute physical risks can act as sources of damage. Chronic risks capture the gradual deterioration of operating conditions associated with climate change. Acute risks reflect the expected damage from extreme weather events, such as floods, storms and wildfires, estimated at the level of individual facilities and then aggregated across companies, sectors and regions. Together, these impacts are used to produce sector-region shocks. These shocks represent the share of productive activity that may be impaired under a given climate scenario and time horizon. They provide the starting point for the supply chain propagation analysis by converting localised physical damage into economic inputs that can be transmitted through production networks, as described in the next section. 

3.2: Propagating disruptions through production networks 

Once sector-region shocks have been estimated, the framework assesses how these disruptions may spread through the economy. This is done using multi-country input-output data produced by the Organisation for Economic Co-operation and Development (OECD)[3], which describe how industries buy from and sell to one another across sectors and regions. 

These data provide a structured representation of production networks. They show, for example, how a manufacturing sector may depend on agricultural inputs, how industrial sectors may depend on mining or energy supply, and how sectors rely on transport, utilities and business services. This makes it possible to identify how a climate shock in one sector or region may affect other sectors that depend on its outputs. A disruption affecting one sector rarely stops there. Reduced agricultural output may constrain food manufacturers; disruptions to mining may affect industrial production; damage to transport infrastructure may affect a broad range of downstream activities. As sectors experience climate-related losses, these losses propagate through supply relationships, affecting customers that depend on their outputs. 

The framework captures not only immediate supplier relationships but also the indirect effects generated through multiple rounds of transmission. A climate shock can pass through several layers of suppliers and customers before reaching a particular sector. As a result, sectors may suffer meaningful impacts even when they are geographically distant from the original hazard and experience no direct physical damage themselves. 

The output of this step is a supply chain damage estimate for every sector in the network. It represents the share of activity impaired by climate-related disruptions occurring elsewhere in the economy rather than at the sector's own facilities. 

3.3: From sector exposure to company risk 

The final step translates sector-level disruptions into company-level exposure. Firms inherit the supply chain risk associated with the sectors in which they operate. Companies with activities across multiple countries receive a weighted combination of the relevant sectoral exposures, following the same geographic logic used for the chronic risk assessment. 

This produces a company-level supply chain damage ratio that is directly comparable with the chronic and acute damage ratios developed in Section 2. Because all three channels are expressed in the same unit, they can be consolidated into a single measure of physical climate exposure and subsequently translated into financial risk metrics. 

The key insight is that a company’s physical climate risk may extend beyond the assets it owns or operates. Physical shocks affecting suppliers, infrastructure and critical inputs can propagate through production networks and materially impair corporate performance even when the company's facilities remain fully operational. By quantifying these indirect effects within the same framework as direct physical risk, the supply chain channel provides a more comprehensive picture of a company's true climate exposure.

This section applies the framework to a diversified portfolio of companies across sectors. The analysis decomposes expected damage into chronic, acute and supply chain components, with the objective of assessing how the different physical-risk transmission channels contribute to overall company-level exposure. 

4.1: Portfolio data 

The framework described in Sections 2 & 3 is applied to a portfolio of 1,444 listed companies operating across 55 countries and spanning 15 Nomenclature of Economic Activities (NACE)[4] codes (A to Q). The distribution of companies varies by sector[5]. Manufacturing is the largest sector in the sample, representing ~42% of all firms. For each company, expected damage is expressed as a share of Earnings Before Interest, Taxes, Depreciation and Amortization (EBITDA) and decomposed into the three transmission channels: chronic, acute and supply chain.  

Facility-level exposures are assessed using the company's geolocated sites. Supply chain exposure is assigned through the company's sector using the propagation methodology described in Section 3.  

Results are reported for the year 2050 under RCP 4.5, a moderate emissions pathway consistent with partial mitigation. This scenario is used to provide a central view of potential exposure under a plausible climate pathway. 

4.2: Study outcome 

Figure 4 reports the decomposition of expected damage by sector and transmission channel. All results are expressed as a share of EBITDA, with acute and supply chain damages reported net of insurance coverage. 

Excluding water and waste management (E), total expected damage ranges from 3.3% in construction (F) to 6.9% in mining and quarrying (B). Sector E records the highest values across all three channels, at 12.3% in total, but it rests on around ten companies, and its results should be read as indicative only. 

Before reviewing the sector-level results, it is useful to return to the food manufacturer example introduced in Figure 1. Manufacturing (C), the largest sector in the portfolio, illustrates the interaction between direct and indirect physical-risk channels. Direct exposure accounts for the bulk of expected damage, with chronic risk at 2.1% and acute risk at 1.8%. The supply chain component adds a further 1.1%, more than a fifth of the sector's total expected damage. For manufacturers that depend on climate-sensitive commodities or intermediate goods, such as the food manufacturer of Figure 1, this share is likely to be structurally higher than the sector average suggests. 

For these companies, the availability of inputs is a source of risk that no investment in the resilience of their own facilities can address. 

Three patterns stand out from Figure 4: 

First, chronic risk represents a common baseline across sectors. Across the fifteen sectors in the portfolio, the chronic component lies within a narrow range, between 1.6% and 2.4%. This reflects the fact that chronic risk is driven by the countries in which companies operate, weighted by their asset footprint. In this portfolio, sector-level geographic exposures are sufficiently similar for chronic risk to form a relatively stable component of expected damage across sectors. 

Second, acute risk varies more materially across sectors. The damage ranges from 0.4% in financial services (K) to 2.9% in wholesale and retail trade (G), with transportation and storage (H, 2.8%), utilities (D, 2.7%) and mining and quarrying (B, 2.6%) close behind. These sectors hold physical assets that are capital-intensive, location-constrained and difficult to relocate, such as ports, rail networks, mines, and power and water infrastructure. Wholesale and retail trade follows the same logic. Its operations run through large distribution centres, warehouses and dense store networks, often located in low-lying logistics hubs close to transport corridors. For these sectors, direct facility-level exposure accounts for a large share of total expected damage, and conventional asset-level assessment captures an important part of their physical risk profile. 

Third, the supply chain channel is material in almost every sector, not only in asset-light activities. It exceeds direct acute damage in many sectors. Even in mining (B), one of the most asset-intensive sectors in the portfolio, indirect exposure (2.7%) is on a par with direct damage (2.6%). This reflects the sector's dependence on energy, heavy equipment and transport services.  

At first sight, information and communication (J) and financial services (K) might appear only marginally exposed to physical climate risk. Unlike manufacturers, utilities or transport operators, they own relatively few climate-sensitive facilities, and a facility-level assessment therefore suggests limited vulnerability. Yet this intuition proves misleading. In information and communication (J), supply chain damage slightly exceeds direct acute damage (1.3% against 1.2%). In financial services (K), the ratio is more than two and a half to one (1.1% against 0.4%). These results highlight that physical risk does not depend solely on the fixed assets a company owns, but also on the sectors it depends on. For example, companies in information and communication (J) often depend on a network of enabling infrastructure and inputs, including electricity supply, data centres and digital infrastructure. Climate-related disruption affecting these inputs or service providers can affect the continuity and capacity to operate, even if a company's own offices and facilities are not directly exposed. The sector's supply chain impact, which matches or exceeds its direct acute damage, reflects the importance of upstream dependencies in assessing physical risk exposure. 

Historical events provide concrete evidence that physical climate risk can propagate through supply chains and critical infrastructure even when a firm's own assets remain largely unaffected. Hurricane Sandy (2012), for example, disrupted the operations of numerous financial institutions in New York through failures in electricity, telecommunications and data-centre infrastructure, highlighting the dependence of the financial sector on climate-exposed service providers and network infrastructure rather than on its own physical footprint[6]. Similarly, the 2011 Thailand floods disrupted a large share of global hard-disk-drive production, generating shortages and price increases that reverberated throughout the information and communication sector worldwide[7]. In both cases, the most significant impacts arose not from direct damage to the affected firms' facilities, but from disruptions affecting upstream suppliers and critical infrastructure. These events illustrate the central argument of this article: a company's climate exposure extends beyond the assets it owns to the broader production and service networks on which it depends. 

Two sectors deserve a note of interpretation: 

Real estate (L) records moderate direct damage (1.2%) despite a business model built on physical property, together with the highest supply chain component in the portfolio outside sector E (2.7%). This reflects the profile of the companies in the sample, which is dominated by real estate managers, agencies and service providers rather than by the underlying properties themselves. Their operations depend on construction, energy, maintenance and facility services. Property-level climate risk is real, but it is better captured by a dedicated real estate framework with its own hazard, retrofit and valuation mechanics. The results for sector L should therefore be read as describing real estate companies, not real estate assets. 

Agriculture (A) records moderate direct damage (1.2%) despite being one of the most climate-sensitive sectors in the economy. This reflects the nature of agricultural value creation. Agricultural operations are exposed to damage affecting land, infrastructure and facilities, but a large share of climate-related losses arises through impacts on biological production systems, including crop yields, livestock productivity and water availability. This sensitivity enters the framework through the chronic channel (1.8%) and through the supply chain channel (1.6%). The supply chain component exceeds direct damage because agriculture itself depends on climate-exposed upstream inputs such as energy, fertilisers, animal feed and water infrastructure. The sector's climate sensitivity is also transmitted downstream, through the supply chain exposure of every sector that relies on agricultural outputs.