From Dependencies to Environmental Risks for Finance. Taking Next Steps

Climate-related risks are now widely recognized as a source of financial risks by financial supervisors and central banks worldwide. Estimates of transition risks, which include risks stemming from the transition to a low-carbon economy, have become more granular, thanks in part to improved data availability. Climate-related physical risks, i.e., the risks stemming from gradual climatic change and weather hazards of increased frequency and intensity, are also moving up policy and research agendas.

Climate-related risks are only one component of the broader category of environmental risks. These stem not only from a higher concentration of greenhouse gases in the atmosphere, but from the degradation of a wide range of ecosystems. For instance, freshwater availability is severely impacted by both climate change and current production and consumption patterns. Yet many economic processes are highly dependent on the regular availability of freshwater. Similarly, declining numbers of pollinators, driven by the use of pesticides and air pollution, negatively impact food production. In addition to these physical risks, nature-related transition risks relate, among others, to the possible economic implications of policies aimed at safeguarding ecosystems (e.g., the expansion of protected areas). This blog’s focus is on research priorities for the assessment of environmental physical risks.

From dependencies to risks

Recent approaches to assessing environmental risks have focused on the dependencies of financial portfolios or entire financial systems on nature. To that end, many studies have used the ENCORE database which assigns a dependency rate for 86 economic processes on 21 ecosystem services. Van Toor et al. (2020) were the first to employ this database in their estimates of the exposure to nature of the Dutch financial system. Looking at the French financial system, Svartzman et al. (2021) combined ENCORE’s dependency rates with a decomposition of companies’ revenue by sector, recognizing that one company may have multiple business segments with heterogeneous exposure. Both approaches included information on the sector-average upstream supply chain from multi-regional input-output tables.

With this focus on dependencies, the discussion has so far revolved around the extent to which businesses are exposed to nature. Further dimensions are needed to move to a notion of risk. Risk is a function of a hazard (i.e., the probability of a harm occurring times its intensity), the exposure to the hazard and the asset’s vulnerability and resilience. Vulnerability can be understood as the degree to which a system is susceptible to damage from the hazard. Resilience describes the capacity of social and environmental systems to react in response to a hazard. In the definition of risk above, the dependency of an economic process on ecosystem services represents the exposure. Information on the qualitative state of the ecosystems in the location of company assets can tell us about its vulnerability. Resilience is captured by the preparedness of the exposed companies to react to damage.

The hazard is the degradation of ecosystem services provided by nature. Forward-looking methodologies need appropriate scenarios of how hazards develop in the future. In that context, current approaches have, e.g., resorted to the assumption that an ecosystem collapses completely (World Bank, 2021; Calice at al. 2021). More appropriate and realistic scenarios should be developed.

Priorities for quantifying environmental risks

To establish workable indicators of environmental risks, we propose four priorities.

First, rather than waiting for comprehensive data to emerge to assess every type of environmental risk, we propose to focus on specific ecosystem services for which data is available. Water-related services would be a good starting point. To improve the estimates of water-related risks, dependency rates can be complemented with existing databases on the qualitative state of water resources across geographies. Georeferenced information on water resources can be matched to georeferenced company data. The environmental risk would then be computed by weighting the dependency rate (the exposure) by the location of the firm’s assets and the state of the ecosystem in that location (the vulnerability) under different scenarios of hazard materialization. In the future, detailed information about companies’ adaptation plans should be used to include resilience in such assessments.

Second, we need to focus on the most exposed supply chains. The next generation of methodologies should determine which sectors are most at risk from degradation of the chosen ecosystem services. To that end, input-output tables indicate which sectors buy inputs from the sectors with the highest environmental risks. The analysis of these supply chains should be prioritized. Building evidence-based, concrete empirical examples for some sectors of the economy allow financial institutions, as well as central banks and financial supervisors to start acting while additional data is becoming available.

Third, this analysis needs to be conducted at the firm-level. Assessments of environmental risks are only helpful for investors and supervisors in the aggregate, if they account for the heterogeneity in exposure, vulnerability and resilience of individual companies.

Fourth, developing more realistic scenarios of losses in ecosystem services and a better understanding of their interaction with economic systems are needed for forward-looking risk assessments.

The role of central banks and financial supervisors

Initial insights into the magnitude of financial dependencies on ecosystems have established a basis for central banks and financial supervisors to consider broader environmental risks, which include risks from climate change and the degradation of ecosystem services. The fact that much remains to be done on the methodologies related to physical climate-related risks should not inhibit financial market actors from moving towards a broader environmental focus. Members of the European Central Bank’s Executive Board have already recognized “the need to further incorporate climate considerations into its policy framework” (Schnabel, 2021). From here it is just a small step to acknowledge the wider implications of environmental degradation.

In addition, further analysis of the overlap of climate- and environment-related financial risks is critical to propel action in both fields. Such analysis would explore, for example, to what extent existing climate risk and ESG indicators can be a proxy for environmental risks. At the same time, progress in the understanding of environmental risks will also improve climate risk assessments, in particular in the context of physical climate risk. Current users of environmental risk metrics should not wait until such metrics cover all risks, but rather start with key risks for which data is available and build up from there.

Central banks and financial supervisors have the means and mandate to develop frameworks to mitigate environmental risks. Although data on environmental risks is in an even earlier stage of development compared to climate risk data, critical insights are already available today. Taking the next steps to translate these insights into risk assessments is vital. As Andrew Bailey, Governor of the Bank of England, said in 2019 on climate risk, uncertainty and lack of data is not an excuse for inaction. The same holds true for environmental risks more broadly.