Sustainability risks & opportunities in the insurance industry

Against the background of insurers’ significant investment volumes, sustainability risks and opportunities on the asset side are of high relevance.

Major issues that can potentially arise from sustainability risks on the asset side include credit risk, market (price) risk, and liquidity risk,¹⁵ as described by, e.g., the German supervisory authority BaFin (2020, p. 18) with the following examples: Credit or counterparty risk can materialize if a company (partially) defaults due to political decisions with respect to ESG aspects that negatively impact the company’s business model such as a carbon dioxide (CO2) tax.¹⁶ Market expectations regarding such political or regulatory ESG measures can also imply market (price) risk of non-sustainable investments by way of depreciation. Finally, liquidity risk may arise in case of a natural catastrophe where a significant number of customers withdraws funds from their accounts to finance losses. Besides climate risks being the dominant issue, further ESG risks, including water scarcity and their impact on assets, are increasingly assessed by financial firms as well (UNEP 2017, p. 41). Other relevant issues or themes are further presented in the Impact Investing Market Map, including renewable energy, education or health (PRI Association 2018a).

While risks can arise from investing in firms that are vulnerable to the consequences of future climate change, for instance, opportunities arise as well when focusing on companies that are particularly resilient to such a development as already emphasized by Herweijer et al. (2009, p. 365). They point out that such an analysis is of high relevance for climate-sensitive sectors, including real estate, municipal bonds and infrastructure funds. An identification of investment opportunities in terms of sustainable firms and projects should also take into account the investment time horizon with a forward-looking approach (Herweijer et al. 2009, p. 365), which is still a valid recommendation today.

Insurance companies might also seize sustainability opportunities by investing in infrastructure projects and by forming public-private partnerships. Swiss Re (2020, p. 23) for instance, estimates “an annual USD 920 billion opportunity for long-term investors over the next 20 years” in emerging markets and expects that “[infrastructure] projects can deliver attractive yields to help insurers match their long-term liabilities, while also offering region and asset class diversification, and opportunity for environmentally and socially responsible investing”.¹⁷ A large amount of financial resources is also needed for the transition towards a decarbonized and climate-resilient infrastructure. Insurance companies are seen as a possible key investor in this context (Golnaraghi 2018).

However, there also exists a number of obstacles which need to be removed first so that insurers can unfold the full potential of their financial capacity, such as complex regulatory restrictions and capital requirements (Golnaraghi 2018), a lack of transparency with respect to data or an insufficient availability of infrastructure investment opportunities in the market (Gatzert and Kosub 2017).

Given that there is a multitude of definitions and concepts, we follow the GSIA (2019, p. 3) as a global sustainable investment organizations’ network by applying the “inclusive definition of sustainable investing, without drawing distinctions between this and related terms such as responsible investing and socially responsible investing” (see also Gatzert and Reichel 2020b). Insurance companies might follow the UN-PRI (or in a broader sense the PSI) in order to apply a sustainable investment approach. In this context, sustainable investment strategies most often include ESG incorporation as an approach, where “asset managers complement traditional, quantitative techniques of analyzing financial risk and return with qualitative and quantitative analyses of ESG policies, performance, practices and impacts” (USSIF 2018, p. 13).¹⁸ ESG incorporation is typically considered to comprise strategies such as ESG integration, negative screening, positive/best-in-class screening and sustainability themed investing. Next to ESG incorporation, engagement (active ownership) represents an additional strategy (PRI Association 2018b; USSIF 2018).¹⁹ Besides implementing (selected) strategies through their own asset manager, insurers might alternatively use external providers including asset management firms or rating agencies (services).

Apart from sustainable investment strategies and asset classes, asset risks can also be assessed by conducting scenario analyses, thereby taking into account, e.g., the recommendations of the Financial Stability Board’s Task Force on Climate-related Financial Disclosures (TCFD 2017b) with respect to transition scenarios and broader physical scenarios of climate change with relevance for both assets and liabilities. Transition scenarios provide “plausible pathways to particular target outcomes” (TCFD 2017b, p. 15) (without statistical likelihoods), e.g. a transition towards a low-carbon economy, a 2 °C scenario, certain CO2 concentration levels, or a net-zero greenhouse gas (GHG) emissions scenario by 2050.²⁰ The process can thus be helpful for firms to understand the time horizon and the potential negative or positive (financial) impact of such a transition towards a low-carbon economy on various industry sectors depending on, e.g., their energy demand and energy prices (BaFin 2020, p. 35). Physical scenarios use outputs of global climate models to focus on the (financial) impact of physical risks from climate change such as from drought or flooding. However, the downscaling of these scenarios to the impact on a regional or local level seems to need further work (TCFD 2017b, p. 24), which also represents a barrier for a broad practical application in insurance companies, as (regional) flood or heavy rain are difficult (if at all) to predict, with resulting uncertainty regarding their impact on firms and thus asset values. Furthermore, the BaFin (2020, p. 35) also cautions that some of the scenarios in transition risk assessment based on IAM were deemed inappropriate.²¹

With respect to climate scenario analyses of the investment portfolio, the UN-PRI supported 2° Investing Initiative offers an open-source and free of charge tool in this context, also known as the Paris Agreement Capital Transition Assessment or PACTA tool. Until June 2020, around 1000 financial firms have already used the online tool in order to assess the impact of a 2 °C scenario transition on their listed equity and corporate bonds investment portfolios, which also allows to align with the TCFD (2017a) recommendations. The initiative also collaborates with several supervisory authorities such as the European Insurance and Occupational Pensions Authority (EIOPA).²²

An application of the TCFD (2017b) recommendations is also conducted by a Working Group coordinated by the Finance Initiative of UN Environment with focus on the banking industry, which resulted in two publications. While the UNEP (2018) report as first part focuses on the assessment of the potential impact of a low-carbon policy and technological transition to mitigate climate change on the credit risk of corporate loan portfolios, i.e. transition risks and opportunities, the second report by UN Environment (2018) also concentrates on risks and opportunities emerging from corporate loan portfolios, but related to physical impacts.

In UNEP (2018), three scenarios with a 1.5 °C, 2 °C, and 4 °C global average temperature increase until the year 2100 are tested in the sense of a sensitivity analysis. To assess transition risks for a firm from the energy sector, for instance, scenarios model its future energy mix along with other variables such as electricity prices, and then the resulting revenues and (emissions) costs among other financial variables. They also compare high vs. low carbon intensity/regulated and unregulated segments. The presented case studies further focus on oil and gas as well as the metal and mining sectors. In addition, a market opportunities assessment is done to identify attractive segments, which is then contrasted with the bank’s capabilities (competitive market, risk appetite, operational capacity) (UNEP 2018, pp. 60–61).

UN Environment (2018, p. 9) continues this work by focusing on physical risks of climate change, taking into account both incremental climate shifts as well as changes in extreme climate events. The impact on the credit risk of agriculture and energy sector portfolios is assessed by looking at sector productivity, revenue and cost of goods sold, and the probability of default, as well as the effect on real estate portfolios (property values, loan-to-value ratios).

In addition or instead of long-term temperature-based scenarios, stress tests can be conducted to assess the effect of sudden changes in policies and technologies on credit and market risks, for instance (UNEP 2018, p. 10). However, as laid out above, the transfer and applicability of scenarios to (e.g. regional) settings and the assessment of the impact is likely highly difficult in practice for many insurers.

In their text mining analysis of annual, sustainability- and investment-related reports and documents of listed European and US insurance companies, Gatzert and Reichel (2020b) observe a strong increase in the number of insurers with a sustainable investment approach over time, especially in the European insurance market. Most important are the application of impact investing and ESG integration in reports and documents, which is also supported by the Geneva Association survey in Golnaraghi (2018) based on interviews with 62 C-level executives from 21 international (re-)insurance companies. Gatzert and Reichel (2020b) find that this ranking has been rather stable over time and does not substantially differ for the considered European and US insurance companies.

However, despite the increasing consideration of ESG issues by insurers, Golnaraghi (2018) concludes that insurance companies are still confronted with numerous challenges on the asset side. The challenges for insurers aiming to support a transition towards a low-carbon economy, and also of relevance for ESG issues, are broken down into five areas (major issues in brackets): financing and market-related factors (lack of general standards and definitions), financial and insurance regulations (regulatory capital requirements), climate change-related policies and regulatory frameworks (uncertainties), technology (risk-return issues in the markets for green technologies) and data and transparency for informed investing (standardization in reporting and stress testing) (Golnaraghi 2018, pp. 21–22).

One key barrier in the investment context thereby represents the first and last area, i.e. the lack of knowledge and data, as ESG data is typically only available for listed (and large-cap) firms and generally not for corporate bonds of small and medium-sized enterprises, which however are relevant for insurers’ asset portfolios. To add to the complexity, whether a business model is “sustainable” or not may also depend on the region (e.g. sufficient water sources etc.). With respect to thermal coal mining or coal power generation, different di- or investment approaches exist in the insurers’ asset management as well.²³ Further problems arise from the question of (out-/under-)performance of these investments, which is strongly mixed (see discussion in Sect. 3.4). Moreover, high costs can arise from ESG integration and other techniques (Golnaraghi 2018), and institutional investors also mention valuation and modeling limitations in the context of ESG integration (OECD 2017, p. 37). Against this background, most insurers use ESG ratings of firms by MSCI (Golnaraghi 2018).

Overall, these challenges might represent a possible explanation that 41% of the 80 largest insurers worldwide still lack a climate-aligned investment approach as highlighted by the AODP&SA (2018) survey. Climate scenario and portfolio emissions analyses continue to be rather in their early stages as well and around 1% of the assets under management are estimated as low-carbon investment.

Besides the UN-PRI initiative, other current initiatives that might intensify the application of sustainable investment strategies on the asset side of insurers and thus increase the future transparency of climate risk exposures, include the UN-convened Net-Zero Asset Owner Alliance, which was initiated in 2019. By aligning investments with the Paris Agreement, i.e. a 1.5 °C scenario, the aim of net-zero GHG emissions within institutional investors’ portfolios shall be accomplished in 2050 at the latest.²⁴ Furthermore, signatories of the Montréal Carbon Pledge, another UN-PRI supported initiative, make a commitment to annually capture, publish and lower their investments’ carbon footprint.²⁵ Also, since 2000, the stakeholder-driven Carbon Disclosure Project supports firms and public bodies (e.g. cities or states) to identify, manage and disclose risks and opportunities resulting from environmental issues with a special focus on climate change.²⁶

On the liability side, climate change plays a particularly important role. For instance, Munich Re estimates 150 billion USD economic losses for 2019 due to natural disasters, of which about one third were insured.²⁷

In regard to such sustainability risks on the liability side, the German supervisory authority BaFin (2020, p. 18) points out potentially adverse changes in underwriting risks with increasing losses in property insurance (e.g., from an increasing number and severity of storms, flooding or hail) as well as business interruption insurance. They emphasize that this should be adequately taken into account in technical reserves, and that these risks cannot only hit the liability side, but also the asset side of the balance sheet. Also, Golnaraghi (2018, p. 7) refers to Carney (2015, p. 6), who lists liability risks in addition to physical risks and transition risks, which may arise from compensation claims made by parties incurring climate change-related damage or loss.

With focus on climate change and weather risks, Herweijer et al. (2009, pp. 362–364) further differentiate between short-term and long-term effects on underwriting practices, as a forward-looking risk perspective with risk-based pricing is already needed in the short-term to account for the fact that historical data is not sufficient for calibrating risk models, which is still relevant today. In the long run, there might even be limits of insurability in regard to affordability and availability of coverage due to correlated risks. Examples named are coastal and inland flooding due to sea level rise along with “super catastrophes” that might increase correlations across regions, business lines and coverages. This is also in line with a UNEP (2017) report, which states that besides natural disasters and climate change, access to and affordability of insurance protection are among the three major sustainability challenges for the insurance sector.²⁸

Opportunities can arise from a potentially increasing insurance demand (depending on the market and type of risk)²⁹ as well as (green) product innovation and microinsurance, that also encourage risk-reducing behavior, and service innovation by way of climate change risk management services, for instance (Maynard 2008; Herweijer et al. 2009; Mills 2009; Stechemesser et al. 2015).³⁰

Besides the product- and service-level, insurance companies can also adapt and benefit through incentivizing low-carbon-related purchases/assets by offering insurance premium differentiation or being transparent on risk and claims performance when underwriting low-carbon assets such as green buildings (UNEP 2017, p. 40). The UNEP (2017, p. 40) report also propagates “to tag claims linked to the asset’s energy performance, fuel efficiency or environmental standards that already exist in a growing number of countries” as a crucial innovation.

Finally, it can be noted that while climate change is certainly the major aspect for the liability side of an insurer’s balance sheet, other sustainability aspects such as social and governance issues might also play a role in D&O policies and reputation risk insurance contracts, and social risks in the context of an ageing population (of relevance for life & pension insurance) are mentioned by the UNEP FI (2009) or Nogueira et al. (2018). Inclusive insurance³¹ or the already mentioned microinsurance policies represent further examples. In this context, the GIZ (2017) highlights the relevant SDGs for the liability side, where six goals³² are identified with insurance having a primary level contribution, i.e. a direct and self-evident impact. For five goals³³, insurance has a secondary level contribution or a rather indirect impact.

To assess sustainability risks on the liability side, (medium- and long-term) scenario analyses can be applied as is done in, e.g., Kunreuther et al. (2013b) in regard to high or low climate change scenarios that address yearly loss rates as well as projections until 2020 and 2040, thereby also taking into account adaptation measures. This is important, as increasing losses can (partially) be reduced by prevention and adaptation strategies, i.e. physical loss reduction measures (Herweijer et al. (2009, pp. 365–366) naming, e.g., more resilient buildings against flood and hazard defenses), which is also relevant concerning the question of insurability. Using scenario analyses for Florida hurricane activity projections and thus future climate variability in terms of high vs. low climate change scenarios, Kunreuther et al. (2013b) provide a first quantitative assessment how the implementation of adaptation measures, also depending on insurance market conditions and reinsurance, influences premiums and thus the affordability and availability of property insurance coverage against losses from hurricane risk. They find that premiums could almost be cut in half in case of fully adopting the physical property-level resistance and resilience measures according to the Florida Building Code 2004, and that a strong and competitive reinsurance market is also vital to ensure availability. However, the authors do not take into account the costs or the climate impact of using a building material other than wood, for instance. An extensive overview on possible measures and solutions with regard to (alternative) risk transfer and reduction in the context of climate change is derived in Stechemesser et al. (2015, p. 565) from the academic literature.

Also of relevance are the recommendations of the TCFD (2017b) as presented in Sect. 3.1, this time with focus on the impact of climate change on the insurer’s liability side (and with the same limitations in regard to scaling to certain regions etc.). In addition to scenario analyses, stress tests should be conducted. The German supervisory authority BaFin (2020, p. 34) thereby recommends to check whether stress tests sufficiently reflect potential sustainability risks for the firm, and to potentially adjust them. References listed include climate-related stress tests for supervisory purposes (i.e. not necessarily suitable for management decisions, but as a first indication for firm-specific analyses) currently developed by the Network of Central Banks and Supervisors for Greening the Financial System (NGFS), the European Systemic Risk Board, the European Central Bank and the German Central Bank.

Finally, it is worth noting that a quantitative assessment of climate-related risks on the liability side and the value of preventive measures and climate policies is highly challenging given the uncertainty and lack of consensus, e.g. in regard to likelihood and outcome (Kunreuther et al. 2013a). As standard approaches are hardly applicable, Kunreuther et al. (2013a) suggest the use of more robust decision-making instruments to assess climate policies where stakeholders have different risk tolerance levels and distribution functions are not (fully) available.³⁴ They discuss the application of different robust concepts using the case of a Florida coastal community as an example that has to decide about whether to allow building a residential facility on the coast. First, as climate models in physical scenarios result in ranges of outcomes, but do not provide entire probability distributions, the authors suggest using a “safety-first” approach where extreme outcomes with very low likelihood below a threshold are not taken into account. Second, in case of ambiguity, where a distribution function is available but with strong variations across different models and thus not objectively defined, as is the case for estimates of the “equilibrium climate sensitivity”, decision-making approaches from the literature can be applied that do not require well-defined probabilities. Third, non-probabilistic approaches such as the minimax regret approach can be used if probabilities of outcomes are entirely unknown.

The results of the Geneva Association survey in Golnaraghi (2018) largely confirm the consideration of the previously presented risks and opportunities for (re-)insurance companies in practice, e.g. risk reduction and transfer, (green) product and service innovation or enhancement and microinsurance. Products tailored for governmental needs and research support represent further relevant measures. External challenges, which hamper (re-)insurers’ contributions against climate change are derived as well, as summarized in Table 2.

In addition to working together with other institutional investors on the asset side in the context of engagement (active ownership) with the aim to address sustainability risks and opportunities at the investee level, insurers increasingly collaborate with other external stakeholders in order to transfer climate knowledge and build resilience against climate change (e.g., UNEP 2017, p. 10; Golnaraghi 2018, p. 18).

As sustainability risks can have severe negative effects on all business units and risk types, from a more corporate perspective, insurers also face various operational risks, strategic risks, and reputational risks (BaFin 2020). Operational risks might occur due to physical climatic events damaging assets as well as affecting the safety and continuity of insurance operations (IAIS 2018).

A strategic risk would arise, for instance, if an insurer fails to adapt to structural changes in the sector, driven by, e.g., the transition to a low-carbon economy and thus not accomplishing strategic goals (IAIS 2018). Furthermore, reputation risk (as a strategic risk) might materialize if the insurer invests in a firm with inadequate safety measures in regard to buildings or working conditions (despite satisfying official requirements), leading to severe and numerous casualties that are reported in the media (BaFin 2020, p. 18). Reputation risks would also arise from false advertising, i.e. greenwashing, or omitting the implementation of “proper” sustainable measures as perceived by external and internal stakeholders (Mills 2009, p. 346; BaFin 2020, pp. 15–18). This reputation risk potential associated with sustainability risks on the asset and liability side led several insurers to divest from coal and ultimately stop insuring the coal mining industry, as was communicated by, e.g. Allianz in 2018.³⁵

On the other side, a reputation reward through sustainable business practices in general is addressed in Herweijer et al. (2009). Based on the identification of sector-specific material ESG or sustainability issues and topics, Clark et al. (2015, p. 13) further point out risk (firm-related risks and external costs), performance (process and product innovation) and reputation (human resources and consumers) as “three major ways how sustainability through the integration of environmental, social and governance […] issues can lead to a competitive advantage”.

Besides diversity on the management or board level (Rao and Tilt 2016, p. 340), workforce diversity³⁶ and inclusion are major sustainability issues, also being considered in the SDGs 5 (“Achieve gender equality and empower all women and girls”) and 10 (“Reduce inequality within and among countries”).³⁷ In addition to promoting corporate sustainability and achieving the SDGs, diversity and inclusion represent opportunities to enhance an insurer’s spectrum of (potential) job applicants, customer satisfaction as well as to thrive innovation through introducing different perspectives, among other aspects.³⁸ In particular, decent places in diversity and inclusion rankings can positively affect reputation.³⁹

Recommendations to establish inclusion and diversity include creating an awareness for diversity, communicate benefits and monitor workforce and board composition.⁴⁰ Insurance companies might also use the methodologies of the aforementioned rankings as a guideline to promote diversity and inclusion within the firm.⁴¹

In general, insurers should further develop a new or adjust their current business strategy to integrate sustainability risks and opportunities (BaFin 2020, p. 19). This also includes the risk strategy and communication⁴² (BaFin 2020, p. 21). With respect to governance, the BaFin (2020, p. 22) highlights the importance of an understanding of sustainability risks and opportunities for the management board, a clear definition of roles and responsibilities, including the consideration in incentive systems and compensation structures (e.g., Fabrizi et al. 2014), as well as the role model function of the management board.

Concerning the risk management in general, Mills (2009) already pointed out that climate change is a typical enterprise risk management issue due to its overarching impact on underwriting, investments and corporate governance.⁴³ Stechemesser et al. (2015, p. 563) further emphasize that the “potentially occurring risks could lead to increasing capital costs, […] increasing operational costs, […] decreasing fixed assets arising from property depreciation” and as a result leading to reduced profitability. Therefore, taking into account liability, transition and physical risks on a corporate level and a result of scenario analyses should be helpful. Mills (2009, p. 352) also suggests to participate in dialogues on public policies concerning climate change, which is in line with Herweijer et al. (2009, p. 371) who suggest that insurers should play a supporting role and have constructive relationships with policymakers, regulators, and other parts of the private sector—all aspects that are still relevant today.

Insurers, irrelevant of their geographic location, can also make use of the US NAIC Climate Risk Disclosure Survey Guidance together with its explanations and “questions to consider” as a reference point for challenging and matching their own climate change strategy.⁴⁴

Based on the aforementioned US NAIC Climate Risk Disclosure Survey from 2012 and 2015, Thistlethwaite and Wood (2018) find that only few insurers integrate a climate change risk management in their investment process and insurance business, with reinsurers being more involved than primary insurers. However, in the Geneva Association report, Golnaraghi (2018) observes that all 21 (re-)insurers included in the international survey consider climate change at least to some extent. While one third treat climate change as an environmental and sustainability issue, 29% are moving forward to increasingly implement it as a core business issue. The remaining (re-)insurers already fully accept and integrate climate change as a core business issue with its risks and opportunities in their strategy and operations as well as in their governance and risk management systems.⁴⁵

With regard to transparency on climate-related risks and opportunities, the AODP&SA (2018) survey finds a discrepancy concerning published information on risks and opportunities by the world’s 80 largest (re-)insurers. While 69% of the sample publish climate-related risk information, only 41% report opportunities resulting from climate change. By applying the TCFD (2017a) framework as a benchmark, the survey further identifies regional differences and, similar to Thistlethwaite and Wood (2018), a disparity among the different insurance business types as well as a lack of appropriate metrics and targets.

In this context, the UNEP (2017, p. 11) report notes that despite large (re-)insurance companies lead the sustainable insurance transformation, a large proportion of insurers lags behind with existing impediments such as different insurance business areas being exposed to environmental risks to varying degrees, misalignment of market incentives or limitations in capacity. The ideal sustainable insurance regime is thereby understood as improving resilience through sustaining and enhancing access, affordability and insurability, reducing risks and exposures through risk management, innovation, prevention, as well as investing in green and robust assets.

With respect to green assets, insurance companies might not only contribute as investors to the greening of buildings and realize higher prices or rents, but also benefit as green building occupants. Reduced operating expenses, improved amenities, employee health and productivity, corporate reputation as well as environment protection represent major benefits and are also seen as possible determinants for the green premium. However, a definite answer whether the benefits outweigh the costs is not possible yet and requires more (empirical) research (Zhang et al. 2018).