- Sector allocation and stock selection can lead to temperature differences between two portfolios. These effects can be measured using an attribution tool, such as a Brinson model.
- As an illustration, we found that a representative emerging-market index exhibited an implied temperature rise of 3.7°C, and a representative developed-market index was substantially lower, at 2.8°C, as of November 2021.
- Systematically assessing the drivers of a portfolio’s temperature allows managers to pinpoint parts of the portfolio where action may be taken to increase alignment with climate targets.
As investors sharpen their focus on the financial impacts of climate change, they may want to understand not only if their portfolios are aligned with global climate goals but why they may (or may not) fall short of the mark. To do so, they may seek to better understand how different sectors and stocks affect how “hot” or “cold” their portfolios are. Investors can do so by turning to tools such as an implied-temperature-rise metric, which projects the temperature of current portfolio holdings into the coming decades.
Previously, we found that approximately 10% of 65,000 individual funds exhibited an implied temperature rise of 2 degrees Celsius (2°C) or less, while 80% fell somewhere between 2°C and 4°C. In this blog post, we extended the analysis to the sector and stock levels, using two representative indexes tracking emerging and developed markets as examples.1
Broadly, we found that a representative emerging-market index exhibited a temperature of 3.7°C and the representative developed-market index was substantially lower, at 2.8°C, as of November 2021. But we can learn a lot more about this 0.9°C gap by delving into what drove this difference.
Understanding the effects of sector allocation and stock selection
We can look at sector allocation and stock selection inside sectors to better understand why portfolio companies may have a higher aggregated temperature than their “cooler” peers. As shown in the exhibit below, sector exposure was an important driver of temperature, since the aggregate temperature by Global Industry Classification Standard (GICS®) sector2 varied strongly between 1.6°C and 6.4°C. Energy, materials and utilities had the highest associated temperatures, which is intuitive, as they were also among the most emission-intensive sectors.
Average temperature by GICS sector
Implied temperature rise of global listed companies by GICS sector, as of Nov. 29, 2021.
Yet company selection within the sector mattered a great deal. Even in the sectors most misaligned with the Paris Agreement,3 we could find companies that are committed to strong emission-reduction targets and therefore compatible with a 2°C world (see the MSCI Net Zero Tracker for more details on companies with strong or weak climate commitments).
Systematically breaking down temperature effects
How can we disentangle these competing effects? To understand the drivers of the 0.9°C difference between the two indexes, we used a simplified Brinson attribution.4 This allowed us to understand if the temperature differences were caused by under- or overweighting specific sectors (an allocation effect) or by having more “hot” or “cold” stocks within a sector (a selection effect).
What drove the 0.9°C difference between emerging- and developed-market indexes?
Brinson attribution analysis for listed global emerging-market and developed-market companies, by GICS sector, as of Nov. 29, 2021.
Using the Brinson attribution, we found that both sector allocation and selection within sectors caused temperature differences. As shown above, the largest heating allocation effects were linked to overweighting the “hot” sectors — energy (0.16°C) and materials (0.09°C) — and underweighting the “cool” health-care sector (0.10°C). The emerging-market index had higher weights in energy and materials than the developed-market index. Additionally, the emerging-market index had a lower weight in health care than its developed-market counterpart, which also increased the aggregated temperature of the index.
For utilities, both indexes had similar weights, so the allocation effect was lessened. But the individual stocks in the emerging-market index had hotter temperatures than those in the developed-market index, leading to a selection effect of 0.10°C for the former index.
When examining the utilities stocks in both indexes in more detail, we found that the emerging-market utilities had, on average, higher emission intensities for Scopes 1, 2 and 3. The higher the emissions of a company, the higher the temperature. In addition, more developed-market utilities (29%) had set emission-reduction targets5 compared to emerging-market utilities (6%), which also helped keep these companies “cooler.”
Aligning with global climate targets
As managers seek to align their investments to global climate targets, they may seek tools to understand why one portfolio is hotter or cooler than another. They can use a Brinson attribution analysis to better discern the drivers of temperature differences between any two portfolios at both the sector and company level. Understanding why a portfolio is hot or cold — or even “just right” — provides important information that can be used in portfolio construction.
1We used the MSCI World Investable Market Index (IMI) for developed markets and MSCI Emerging Markets IMI for emerging markets. All constituents are as of November 2021.
2GICS is the global industry classification standard jointly developed by MSCI and S&P Global Market Intelligence.
3The 2015 Paris Agreement seeks to keep global temperature rise, by the end of the century, to well below 2°C from preindustrial levels.
4The calculation is analogous to the Brinson attribution technique that allows managers to break down a portfolio’s active return into an allocation and a selection component. One simply needs to replace a stock’s return by the stock’s implied temperature rise and the stock weight by financed-emission-budget weight in the formulas. Note that a more detailed approach explicitly shows the effect of the interaction between the allocation and selection components, which is generally small. Here, for the sake of simplicity, we absorbed this interaction term in the selection component. The methodology applies to multi-asset-class portfolios as well.
5A company that pledged to reduce its emissions is projected to emit less over time, and therefore its implied temperature rise is also lower.