Idea #18

Redirecting the Capital Flows of the Insurance Industry

by Alexander Braun

Current estimates show that the average temperature on earth has already increased by one degree centigrade since the industrial revolution. Anthropogenic emissions of greenhouse gases, such as carbon dioxide, are the major driver of this development, causing temperature anomalies that persist for millenia.  Permafrost thawing, extreme weather patterns, ocean acidification, polar cap melting, and desertification are just a few of the irreversible consequences earth could be faced with if mankind fails to take quick and decided action.

Fortunately, during the most recent conference of the parties (COP) held 2018 in Katowice, Poland, nearly 200 nations approved obligatory rules for the measurement and reporting of their efforts to mitigate climate change.

Clearly, a substantial transformation of power generation and manufacturing infrastructures will be necessary to achieve the long-term temperature goal of 2°C. To accelerate this process, global capital flows should be redirected towards low-carbon technologies. The insurance industry could play a key role in this regard, as both an investor and an insurer.  This is owed to the classical insurance business model, under which premiums collected from policyholders in exchange for coverage are not kept idle but are put to work in the capital market. This generates an additional source of income for the insurer and is possible, since payouts for insured losses typically occur with delays. Consequently, insurance balance sheets essentially consist of two portfolios: the investment portfolio, which forms the asset side, and the underwriting portfolio, which, together with the equity capital, represents the liability side.

Recognizing their responsibility, numerous insurers and reinsurers have already been proactive and started climate-related engagements such as financing mangrove reforestation, advancing loss prevention, and promoting disaster-resilient and energy-efficient building practices. Mills (2012), for example, discusses 1,148 initiatives from 378 insurance companies in 51 countries. Furthermore, 65 entities adhere to the Principles for Sustainable Insurance (PSI), a voluntary framework introduced in 2012.

Yet, it is unclear how effective these efforts really are. Some firms might simply be window dressing through small-scale investments in green technologies or climate-friendly funds, taking advantage of the positive reputation effects associated with sustainable business policies. A genuine impact, in contrast, will only be achievable if insurers strive for strict carbon-neutrality of their investment portfolios. This means that they should refrain from investing in the stocks and bonds of companies which generate high levels of direct carbon emissions. The latter are typically from the power production, heavy manufacturing, and transportation industries. Anecdotal evidence indicates that such a change in investment philosophy would likely not be associated with a sacrifice in terms of expected returns. Recent empirical results even point to the possibility of a low-carbon premium.

The potential is huge: estimates for the for the global insurance sector range close to  USD 25 trillion in assets under management, which is more than 15-fold the projected private sector gap that needs to be closed to achieve all 17 United Nations sustainable development goals (SDGs) by 2030. A mere partial redirection of this capital could be a substantial accelerator for the transition to a low-carbon economy. As extant sustainability principles are not mandatory and due diligence is associated with information costs for stakeholders, however, there are no strong incentives for insurance companies to free their balance sheets of carbon exposure. Barring a current study by Mielke (2018), the scholarly literature is surprisingly quiet on this matter.

In a forthcoming research article, my coauthors and I propose a new climate-change policy for the insurance industry, consisting of two main elements. First, we harness asset pricing theory to design a rapid test for carbon exposure in the investment portfolios of listed insurers. At the heart of the approach is the EU Emissions Trading System (EU ETS), which limits the total amount of emissions in the economy and requires polluters to purchase greenhouse gas (GHG) certificates, forcing them to internalize the external effects caused by their activities. Since this mechanism puts a price on CO2 emissions and turns them into a scarce resource, it affects the operating costs of carbon-intensive businesses. Consequently, rises in the price of CO2 should decrease the expected future profits of polluting companies and, in turn, the prices of their stocks. If this effect exists, it should also leave traces in the stock returns of insurance companies, who hold such assets in their investment portfolio.

To detect potential CO2 price exposures hidden in insurance balance sheets, we thus suggest extending an existing factor model for insurance stocks by the excess returns on European emission allowances. Through an empirical estimation of the coefficient for this additional “carbon factor,” it should be possible to objectively measure the carbon intensity of the constituents of the insurers’ investment portfolios. This is important, since actual investment practices may diverge from proclaimed intentions, a problem known as style drift in asset management. In other words, a company might well announce a rebalancing of its asset base to low-carbon positions, but then not fully deliver on its promise. Based on a sample of 35 European insurers, we illustrate the implementation of the model and analyze the time-varying patterns of the carbon factor coefficients (betas). Due to the supposed negative relationship between the price of CO2 and the stock prices of heavy emitters, higher betas imply less carbon exposure. Most firms exhibit an observable increase in the carbon betas throughout 2018, which is in line with explicit public declarations to decarbonize their balance sheets.

Second, we suggest a number of accompanying regulatory changes. One key aspect concerns the institutionalization of the suggested carbon test and the consequences for firms which exhibit a significant CO2 coefficient. A straightforward way to tackle this question is an integration into Solvency II. Insurers could, for instance, be obliged to publish the carbon beta in their annual report and their Solvency and Financial Condition Report (SFCR). Stakeholders of the firm would thus have an easy and inexpensive way to evaluate the firms’ climate compatibility. Moreover, the results of the suggested carbon test could be utilized for the introduction of an environmental, social, and governance (ESG) label for insurers akin to existing signals and rankings in the investment fund industry. Finally, as a measure of last resort, regulators could contemplate a rebate in the capital charges for insurers with green balance sheets and a markup for those with significant carbon exposures. While an adjustment of risk-based capital standards based on mere political considerations is certainly debatable, increasing carbon regulation and investor scrutiny could indeed change the risk profile of heavy emitters in the medium to long run. After all, empirical research has already documented a comparable effect for the stocks of companies in the tobacco, alcohol, and gambling businesses.

As this climate risk solution is still in an early stage, further research is needed to mature the concept. In particular, the suspected link between GHG emissions and stock returns needs to be better understood. Although equities constitute only a minor part of the typical insurer’s investment portfolio (current estimates for the US amount to 12.5%), the potential to drive the shift to the low-carbon economy is still great: as mentioned above, total industry assets are estimated around 25 trillion worldwide. In addition, the relationship between CO2 exposure and bond returns should be investigated. About 60% to 70% of the investment portfolios of insurance companies consist of government and corporate debt, implying a massive lever if one has a model that can detect bonds of polluters.  Finally, the liability side of insurance balance sheets is not covered by the above-mentioned approach. This means that firms may appear to be climate friendly, since they run a low-carbon asset portfolio, although they still insure CO2-intensive facilities, such as coal plants. Hence, to mobilize an even greater capacity for the mitigation of climate change, the capital flows of both the investment and the underwriting portfolios must be redirected. Although there are no straightforward solutions to these issues, considering the high stakes involved, their further consideration is well worth the effort.

This post is based on the article, Alexander Braun, Sebastian Utz & Jiahua Xu, “Are Insurance Balance Sheets Carbon Neutral? Harnessing Asset Pricing for Climate-Change Policy”, forthcoming in the Geneva Papers on Risk and Insurance, which won the 2019 Shin Research Excellence Award of the Geneva Association and the International Insurance Society (IIS).

Alexander Braun is Visiting Scholar at the Wharton Risk Center.


Home