Asset class considerations
Next, we touch on several key asset class considerations that asset owners will need to factor into their implementation decisions.
Active vs passive
As noted earlier, there are a variety of climate investing strategies, including both active and passive strategies. One of the advantages of active strategies, in our view, is that managers can take multiple approaches in pursuit of the investment objectives: engaging with companies, reducing a portfolio’s weighted average carbon intensity (WACI) or the implied temperature rise (ITR)1, or investing in climate solutions, for example. In addition, active managers can use detailed research to help uncover value or find inconsistencies in reported data.
Passive allocations can also address climate across multiple dimensions (e.g., through the use of Paris-aligned benchmarks2), but this approach will be less nuanced, less able to evolve, and potentially more exclusionary. For example, passive funds tracking a Paris-aligned benchmark typically aim to minimize WACI or ITR. Since the indices themselves are designed to minimize tracking error versus a traditional index, this more statistical approach may miss out on opportunities that active managers can pursue.
We delve deeper into the active/passive decision later in the paper.
Public vs private assets
While the bulk of the climate investing universe has historically been focused on the public market, private asset approaches have grown meaningfully in size and maturity. We think private equity can be an effective way to tap into innovation. Here, climate investments can be either early stage and more “science focused” (e.g., on technologies that have yet to scale, such as direct air carbon capture) or later-stage, more proven technologies.
It is worth noting that climate-related metrics for private assets are currently fairly limited compared with the public market, so it can be challenging to quantify or assess the climate intensity or positive impact of private investment approaches. One solution could be to use public assets as a proxy for private assets. In addition, when possible, the disclosure of climate goals should be an objective of engagement when investing in private assets, especially given that the ownership structure provides the investor more influence (often, information may be available but not disclosed). Forthcoming climate disclosure rules in the European Union (EU), Australia, and California require disclosure from companies — listed or private — that meet certain revenue and other thresholds, which could improve data availability. At Wellington, we have been involved in the ESG Data Convergence Initiative (EDCI), which seeks to create a set of comparable and meaningful ESG data (including in climate) from private companies.
Sovereign bonds
One of the biggest data gaps exists within sovereign bonds, where asset owners and managers have historically not applied climate investment frameworks. Here, the issue isn’t one of disclosure, as both production- and consumption-based emissions metrics are widely reported, as is other data such as that on fossil fuel exports. However, investment frameworks are at an earlier point in their evolution, with initiatives such as the Assessing Sovereign Climate-related Opportunities and Risks (ASCOR) Project aiming to develop assessment methodologies for sovereign bonds. At Wellington, we have a framework for assessing sovereigns on climate metrics that considers developed and emerging markets separately and assigns a greater weight to transition metrics for developed markets. For emerging markets, physical risks are key. We also engage with Woodwell Climate Research Center on regular climate “deep dives” on particular countries to enhance our understanding of country-specific challenges and opportunities.
Other tools can also be leveraged for assessing sovereign portfolios. For instance, as noted earlier, the NGFS publishes climate transition and physical risk scenarios. They could be used to evaluate countries’ current policies and climate-action plans (Nationally Determined Contributions or NDCs3), as well as their impact on macro variables such as GDP and inflation, and on what achieving carbon neutrality by 2050 could mean for country fundamentals.
Reconciling data issues and data availability
In our last paper, we took an in-depth look at the addition of climate-related metrics to the strategic asset allocation optimization process and the ways in which that process can help with multiple objectives — for instance, a better WACI and a better ITR. But what about trade-offs within asset classes? Following are a few thoughts on navigating the data limitations discussed above:
Private assets — When data for the optimization process is lacking, we think the answer is not to penalize the asset class. Rather than reduce or avoid allocations to private assets on the basis of data availability, other levers can be used at the implementation stage (e.g., engagement).
Public assets — For public assets, where data comparability can be an issue, the optimization process can be conducted at the gross asset class level — e.g., improve the ITRs separately within each asset class. We find this is generally possible within equities and credit, while for sovereigns, a lack of ITR data may require a different approach — looking instead at sovereigns’ current policies or NDCs, for example. However, as noted in our last paper, this can result in high turnover and deviations from the benchmark.
Company emissions — With respect to data availability for company emissions, Scope 1 and 2 data is generally readily available, but access is more challenging as the data becomes increasingly nuanced.4 With Scope 3 data and physical risk data, it becomes necessary to rely more on company reports, and issues of comparability can emerge. As noted, ITRs can be useful in this exercise and are forward-looking, but they may also rely on more opaque calculations from third-party providers. Other forward-looking data considerations include whether a company has committed to decarbonization targets under the Science Based Targets initiative (SBTi) and whether those targets have been validated.
Impact metrics — We have an internal impact measurement framework that seeks to provide as much consistency as possible in each sector, but again, the information companies provide is not always comparable. As a result, it may be necessary to use “adjacent” KPIs, such as data on the percentage reduction in energy use (versus emissions avoided). Based on our experience, it is imperative to avoid false precision — i.e., don’t make assumptions if it’s not necessary.
Physical risk — Physical risk is challenging to evaluate, but there are tools that can help. As an example, our Climate Research Team has created a tool that uses location data for more than 10,000 public companies and several thousand securities. The team has evaluated the physical risk of about 92% and 71% of the market value of the S&P 500 and MSCI ACWI, respectively, as of the end of 2023.
For sovereigns, the NGFS scenarios, which have been enhanced with more robust modeling of chronic and acute physical risk, can be leveraged. They may enable a holistic view of a portfolio should the net-zero targets fall short and physical climate-change risks accelerate. From an alpha potential standpoint, we think it will be important to be exposed to countries with the ability to adapt and with the lowest impact on GDP from climate change.
Data availability for different issuers — To provide a sense of the tools and data available across issuers, Figure 1 shows some of the key carbon footprint metrics discussed for a number of equity and fixed income indices. Not surprisingly, there is less coverage for the high-yield fixed income universe, where issuers are further down the market-cap spectrum. The table also highlights the different WACI methodologies investors should be aware of when looking at corporate versus sovereign indices. Within the sovereign space, the choice of the production method or the consumption method of calculating carbon intensity can influence the output and portfolio-level exposures. Being aware of the choices and assumptions behind each metric used to assess a carbon footprint is important, especially if the metrics will be used to inform allocation decisions or as an input to any optimization process.