Climate-Related Infrastructure Failure Has Complex and Far-Reaching Economic Impacts

The full extent of climate change impacts is currently not integrated into financial risk assessments or valuations. Even with a raft of new regulation, impact assessment methodologies are left wanting with little emphasis on physical risk beyond an asset focussed approach.

However, extreme climate events can cause significant business interruption which may be independent of any direct damage to assets a business may suffer. This means that economic losses from future climate tail events are likely to be heavily underestimated, especially for institutions with exposure to cash flows of the companies they lend to or invest in, as opposed to asset values alone.

Indirect Catastrophe Impacts May Be Large

The rapidly escalating frequency and intensity of heat waves, floods, hurricanes and other such climate events is creating novel challenges. Critical infrastructure, especially older infrastructure, may not be designed for current, and future, catastrophe extremes. Major climate events will likely overwhelm critical infrastructure, including economically vital systems and networks such as energy, water and transport, with knock-on impacts on education, health and public security.

Sahil Shah

Macroeconomic models demonstrate the importance of energy and water inputs, especially for the secondary sectors, including manufacturing, the chemical industry and construction; they are also a requirement for the tertiary sectors to function (e.g. hospitals, schools). Moreover, transport disruption will likely prevent workers from traveling to their workplaces, creating major shocks where remote work is not possible or planned. These disruptions have, in fact, already occurred in major climate events such as Hurricane Katrina, Hurricane Harvey and the 2020 South Asian Floods.

These extreme climate events can cause significant interruption independent of any direct damage to a business’s assets. For example, a hotel on a Caribbean island may suffer extensive business losses due to a hurricane that damages the local port, preventing tourist arrivals, irrespective of whether the hotel itself is damaged, as shown in Figure 1.

Banks are particularly sensitive to loss created by damaged infrastructure, as asset impairment alone will not necessarily result in borrower non-performance. However, cash flow impairment from business interruption, if severe, would likely force borrowers into default. Consequently, default rates could soar in the wake of a severe climate event that’s insufficiently addressed in risk assessments. Such an outcome could be particularly problematic for banks with geographic or sectoral concentrations.

Edward Baker

As climate hazards become more severe, losses and credit failures from indirect disruption impacts may exceed those from direct impacts. Understanding this point is imperative for lenders and investors as it gives them a strategic edge and can help to reduce potential losses. A deep understanding of how climate hazards unfold, when infrastructure fails, how this affects transport and production and translating this back to cash flow impacts creates an opportunity to better manage the financial impacts of these risks. The most forward-thinking lenders and investors will see this as an opportunity, where they can aid borrowers and investees in their stress testing and climate resilience plans and thereby offer differentiated pricing, rewarding those best prepared.

Current risk toolkits and methodologies, such as those published by the Network for Greening the Financial System (NGFS), focus on transition risk, and the physical risk that is included is at the asset level with scant consideration for indirect impacts that affect economic activity in a much more disruptive sense. An approach that examines the indirect impacts is therefore needed, especially as basic correlation analysis shows increases in credit failures after major natural hazards, across sectors, including those without physical assets.

Understanding Climate Change’s Indirect Impact on Credit Risk

To better understand how climate change affects credit risk through indirect channels, we propose the following methodology.

1. Selecting the most important critical infrastructure types to take into account. Transport, energy and water systems are vital for economies to function and loss of these creates major challenges for businesses. 

2. Identifying the impact channels. Depending on the infrastructure types and hazard exposure, impact channels will vary. Below we list the impact channels for the most critical infrastructure types.

   Transport:
   • Floods can lead to road closures, impede vehicle speeds, damage port facilities, force closure of rail/underground stations and damage railway/subway lines.
   • Cyclones and hurricanes exacerbate flood damage through wind, storm surges and wave impacts, potentially damaging ports, power substations and breaking transport pipelines.
   • Heatwaves have the potential to damage transmission lines for railways and overheat transformers; at extreme levels they can melt tarmac.
   • Fires cause direct damage to railway subway/transmission lines, stations and ports.
   • Droughts can affect riverside ports through reduced water levels, thereby impeding anchorage.
   • Importantly, these transport disruptions can cascade into other critical infrastructure where key workers are unable to get to workplaces.
   Energy:
   • Heatwaves have the potential to damage electricity transmission lines and overwhelm transformers.
   • Floods directly damage power substations, and oil and gas production.
   • Transport disruption from floods also impacts trucks and ships carrying oil and LNG.
   • Excess wind speeds from cyclones, hurricanes and tornadoes can damage oil and gas pipelines.
   • In extreme circumstances, water shortages from drought can lead to the closing of power plants and other water-intensive energy infrastructure, such as oil/gas production. Alternative energy sources are not immune to these disruptive impacts; for example, certain types of lithium mining are highly water intensive and forced water use restrictions can greatly impact mine output.
   Water:
   • Impacts on dams, water trucking and water pipelines all occur through similar transmission channels as outlined for transport and energy.

3. Modeling infrastructure breakpoints. Assessing infrastructure breakpoints is complicated. Chronic degradation over years due to weathering, weft/warp strength of construction materials used, and design of infrastructure (e.g. how components are joined together) are all factors that are important to assessing infrastructure breakpoints. Building bespoke fragility curves layer by layer that increasingly factor in the nuances of infrastructure vulnerability enable accurate prediction of potential infrastructure failure, and therefore the likelihood of failure under particular climate events.

4. Calculating economic disruption. Macroeconomic models such as input/output models, dynamic stochastic general equilibrium and agent-based computational general equilibrium models can be used to convert identified breakpoints into input shocks that subsequently lead to drops in output and price rises. Such modeling can be supported by other forms of analysis. For example, network analysis can be used to model disruptions in transportation networks and to show changes in network capacity and rerouting.

5. Modeling the post-disaster recovery time. Infrastructure capacity recovery times can vary greatly, from days to months. Historic data from previous disasters and other locations that are similar (economically, socially, culturally or other) can inform these time estimates.

6. Incorporating impacts into portfolios. The economic disruption assessments and estimates of time to recover post-disaster can then be incorporated into a stress testing framework for evaluating cash flow impairment under varying assumptions about the magnitude of the hazard, yielding a distribution of possible business impairment outcomes. Lenders will then be able to incorporate the possible cash flow impacts from climate tail events into their credit risk assessments.

Incorporating the Impacts Case Study

Let’s return to the example of the tourist hotel on the Caribbean Island. Suppose the hotel is interested in adding a wing and it approaches its bank looking for a five-year-term loan of $50 million. The bank is, of course, concerned about the impact of hurricanes on the hotel's ability to pay back the loan. They have two concerns — first the direct impact on the hotel and its property, second the indirect impact on tourist arrivals resulting from damage to the island’s only port. The port is old and susceptible to damage in extreme wind, rain and storm surges and would likely be inoperable for an extended period if an extreme hurricane occurred.

The impact of the hurricanes would be during the peak summer months when hotel occupancy is normally 90% or more, and this is the period when the hotel generates the bulk of its income. The bank considers the possible impact of three hurricane event scenarios: 1) Hurricanes of average intensity, 2) Intensity in the top 25%, and 3) Intensity in the top 5%.

A thorough risk assessment determines that a hurricane of average intensity would cause moderate damage to both the hotel and the port with occupancy rates dropping to 75% due to the direct damage and 80% due to damage to the port, with damage to both repaired in a few weeks. A top 25% event would likely cause a drop in occupancy to 60% due to direct damage, but 50% due to the expected port damage, with the hotel and port damage repairs both taking a couple of months. However, a top 5% hurricane would reduce occupancy due to direct damage by 50% lasting about two months, but the port damage would be extensive reducing occupancy to 20% and would take at least six months to repair (see Table 1).

The bank further assesses that the likelihood of extreme 5% historical events has increased in recent years and their intensity has been rising. Based on this observation, and the assessment of the indirect impacts on occupancy rates from the three scenarios, it decides that the potential loss in the hotel’s revenues in the extreme scenario would likely result in the hotel not being able to service the loan. Therefore, it decides not to lend to the hotel.

Parting Thoughts

Climate-related risks are messy, complicated and challenging to predict, and the indirect impacts can be large. Understanding and modeling the cascading impacts from damaged infrastructure provides a commercial opportunity not just to better assess and manage risk, but for financial institutions to provide preferential terms for borrowers more prepared for climate impacts. This has the ancillary benefit of providing clear commercial objectives for businesses to become more climate resilient.

A deep understanding of climate risk impacts on business disruption enables differentiated loan pricing strategies that incorporate the climate/credit risk nexus at a loan level to better price climate change-related risk, and the smartest investors and competitors will understand this opportunity.

If you would like to learn more about this topic, you can listen to a recent GARP Climate Risk podcast with Emma Howard Boyd CBE, discussing cascading and interconnected risks in cities.

Sahil Shah is co-founder of climate risk modeling firm Tipping Frontier. He has been a fellow at the Atlantic Council, an honorary research fellow at the University of Madison-Wisconsin, a member of the University of Cambridge’s Global Food Security Network. He has fed into National Adaptation Plans and trained global financial institutions in climate risk modeling. He holds an MA in Economics with Management from the University of Cambridge.

Edward Baker is an advisor to climate risk modelling and adaptation finance firm Tipping Frontier. His career spans 40+ years in the asset management industry in diverse roles — starting as a founding member of risk management modelling pioneer Barra Inc. He earned an MS degree in mathematics from the University of California at Berkeley as a Regents Fellow and worked in PhD programs in both mathematics and finance.