Liability-driven investing and climate risk: facing reality one step at a time

In liability-driven investing (LDI), in addition to unrewarded risks, default risk and migration risk are the most significant. One of the main goals of an LDI strategy is to ensure that expected cash flows are received in order to meet the pension plan’s liabilities or, in other words, minimize default risk. Migration risk arises when a company’s credit rating falls below BBB, thereby becoming a high-yield issuer. By exiting the investment-grade universe, the impact on the performance of an LDI strategy is significant, since the manager would, in many cases, be forced to sell the company’s bonds under conditions that aren’t necessarily favorable.

Maintaining a good credit rating is therefore crucial, and while factors such as having a healthy balance sheet and profit margins remain fundamental to issuers’ credit analysis, the impact of climate change now adds a new dimension to migration risk.  

Understanding climate risk

Climate risk comes in three forms for companies:

Source: Manulife Investment Management, Office of the Superintendent of Financial Institutions.

Climate risk is systemic (i.e., undiversifiable) and will affect nearly every company around the world; however, the magnitude of the impact of these types of risks/opportunities will depend on how we approach climate change as a society. For instance, the effect of the transition factor will be higher in a scenario in which we succeed in limiting the temperature increase to only 1.5°C by 2050, as this will mean that significantly stricter regulations will have been put in place to achieve this.

In our view, the most likely scenario is one of those provided by the Network for Greening the Financial System, which assumes that global temperatures will increase by 2°C by 2050 and that the transition will be disorderly across sectors and regions. In this scenario, our society is somewhat slow to act and therefore has to accelerate later to rectify the situation—a reactive rather than preventive approach. This scenario suggests that the transition component will have a relatively high, but longer-term, impact, meaning that companies have somewhat of a grace period before more drastic policies are implemented.

In an LDI context, another important factor to consider is the capital structure. Since we invest in bonds, which have a higher priority in the capital structure than equities, the climate shock is milder. The effect of a new policy or climate event is initially absorbed by a decrease in equity value, and depending on its financial strength, its bonds and credit rating could remain intact.

Furthermore, in the event of bankruptcy, bondholders have some protection, as they're among the first investors to be reimbursed from the disposition of assets. The average recovery rate frequently cited in the industry is 40% in the event of bankruptcy; loss of value of a bond should therefore be between 0% and 60%.

A company with a strong balance sheet can better absorb a climate shock

Source: Manulife Investment Management, MSCI, 2022. For illustrative purposes only.

Measuring climate risk

Several providers of climate risk data and measures are currently available on the market, and we’ve considered and analyzed several of them up to this point. This includes the raw data associated with GHG emissions, whether in absolute or intensity terms, which can be broken down based on the value chain, from upstream direct emissions in scope 1 to indirect downstream emissions in scope 3 and energy consumption emissions in scope 2. One of the limits of this data, which is nevertheless tangible, is the fact that it’s historical and doesn’t make it possible to see how a company manages its climate risk, which could have an effect on its future valuation and consequently on an LDI strategy. Also, this type of data doesn’t necessarily capture all forms of climate risk such as technological opportunities.

For these reasons, we’ve chosen to work with a forward-looking measure that considers the three previously described forms of climate risk and is in line with the recommendations of the Task Force on Climate-related Financial Disclosures for analyzing scenarios in investment portfolios. The climate value-at-risk (climate VaR) metric provided by MSCI aims to measure, in percentage, the potential negative impact of the effects of climate change on the valuation of issuers (at the company or share level) or individual holdings (at the security level) based on different scenarios.

As of November 30, 2022, the climate VaR methodology coverage of corporate issuers was 92% in the United States based on the Bloomberg U.S. Corporate Bond Index.¹ In comparison, this figure is 73% for Canada based on the FTSE Canada Universe Index.¹ This data is also available at the level of most bonds, providing additional granularity as they take into account their maturity. Climate VaR is calculated using raw, quantifiable underlying data, such as GHG emissions, and uses a rigorous methodology, providing investors with relatively accurate and reliable data. It’s worth noting that we’re continuing to analyze and validate with our sustainable investment team the change in different methodologies from different data providers. Even though this data is separated into the three types of climate risk (physical risks, transition risks, and technological opportunities), the physical risk factor gives results that are too limited to be included in our analyses at this time. Within our first case studies for integrating climate VaR, the following conditions are therefore based only on transition and technological opportunity factors. 

Conclusion 1: U.S. bond markets display lower long-term climate VaR than Canada

We assessed the two biggest bond markets in North America—the United States and Canada—to compare the level of climate VaR with what an LDI mandate would likely encounter. At first glance, the Canadian corporate bond market is less exposed to climate risk, with a total climate VaR at 6.0% compared against the United States at 6.6%.² In addition, Canada has only two sectors that could be considered high risk, while the United States has four.

Bond climate VaR, by sector and in total, weighted average (%)

Source: Manulife Investment Management, MSCI, as of November 30, 2022. REITS refers to real estate investment trusts.

However, on closer analysis, managing corporate bond climate risk is less complex in the U.S. market than in Canada. This is because more than 35% of Canadian corporate bonds are short-term bonds from the financials ex-insurance sector, which significantly reduces the total climate VaR of the Canadian market since these short-term bonds have virtually zero exposure to climate risk. LDI mandates invest primarily in mid- to long-term bonds to match the duration of liabilities, which is typically between 12 and 25 years; therefore, many of the Canadian short-term corporate bonds aren’t effective at reducing a mandate’s climate risk.

A look at the longer-term segments, which better represent the investments of an LDI mandate, shows that the U.S. market is actually less exposed to climate risk than the Canadian market where there’s a higher proportion of long-term bonds from oil and pipeline companies compared with the United States.

Climate VaR for longer-term corporate bonds in an LDI context (%)

While there are other factors to consider, such as basis risk and meeting investment objectives, we believe the U.S. bond market is a good option for diversifying the climate risk exposure of LDI mandates. 

Conclusion 2: Security selection has more impact than curve positioning

Naturally, there's a direct relationship between the term of a bond and its exposure to climate risk (all else equal, the longer the term, the higher the climate VaR). However, we would have expected the difference to be more pronounced, especially in a scenario in which the temperature increase is limited to 2°C. For instance, the difference between two bonds from the same U.S. retailer—one of which matures in 20 years and the other in 29 years—is only 2%. Meanwhile, the difference in the climate VaR between this same 29-year bond and a bond of a similar retailer with the same term is 13%.

Examples of the distribution of climate VaR for medium-/long-term bonds from two major U.S. retailers

These results indicate that, in order to mitigate climate risk, selecting companies with better environmental practices is more effective than investing in a bond of the same company, but with a shorter maturity. When carefully executed, security selection and sector allocation can reduce climate risk without affecting interest-rate sensitivity, unlike a shift in curve positioning.

Conclusion 3: Climate VaR data leaves room for judgment

It’s worth noting that the climate VaR data is based solely on the quantitative data available at a specific point in time and doesn’t take into account the potential and commitment of an issuer to improve its environmental practices. In our view, two identical climate VaR numbers aren’t necessarily equal, and they necessitate human judgment. Fundamental research and commitment from issuers will always be a necessary addition to any quantitative approach to nuance some results. For instance, the following two issuers each have a climate VaR of 60% (the worst possible climate VaR), but when we look at the investments made by both companies in recent years and their climate plans, it becomes clear to us that Company A’s climate risk is lower than Company B’s.  

The next steps

As demonstrated above, we believe climate risk is now an integral source of migration risk, and while the use of Climate VaR data provides a good starting point to better identify and manage this climate risk, our work doesn’t stop there.

Last, the environmental aspect often receives the most attention among the three ESG factors, but the other two factors—social and governance—are never far behind. Are we soon going to see a social and governance VaR emerge, especially when we look at other systemic issues, such as human rights and inequality when it comes to human capital? What’s certain is that the level of complexity in analyzing and managing corporate bond migration risk will continue to increase, so the ability to establish a rigorous and meticulous credit analysis process has become more important than ever for LDI strategies.

1 The Bloomberg U.S. Corporate Bond Index tracks the investment-grade, fixed-rate, taxable corporate bond market. The FTSE Canada Universe Index tracks the performance of marketable government and corporate bonds outstanding in the Canadian market. It is not possible to invest directly in an index. 2 That means a decrease in value of 6.0% in a portfolio containing Canadian investment-grade bonds versus a loss of 6.6% in a U.S. bond portfolio.

Climate value at risk (climate VaR) is designed to provide a forward-looking and return-based valuation assessment to measure climate related risks and opportunities in an investment portfolio.

The climate VaR model developed by MSCI aims to measure the potential impact of different climate scenarios on individual securities’ valuations. Climate VaR indicates, in percentage points, what could be the potential impact on a security’s market value as a result of the effects of climate change.