Climate change and carbon emissions matter, but so does the cost of managing these – particularly in a concentrated investment universe like the JSE. We evaluated the effects of managing carbon emission on the purity of the portfolio factor signal (value, momentum and quality) and risk-adjusted returns by either excluding the most carbon-intensive companies or creating portfolios that achieve a less-than-benchmark carbon-intensive level. The results are clear: it is possible, but the impact on factor scores and risk-adjusted returns is consistently negative. These findings suggest that it is more beneficial and impactful for South African investors to look for these gains in their offshore portfolios to reduce their carbon footprint holistically. They should encourage their investment managers to engage with the management teams of the most carbon-intensive companies to get them to focus on carbon emissions reduction targets and the strategies necessary to get them there. |
Regulators in South Africa and the industry-developed (voluntary) Code for Responsible Investing South Africa (CRISA) also require investment managers to incorporate ESG principles into the responsible management of their clients’ assets. However, it is also important to recognise that any restrictions on a portfolio will impact its risk and return profile.
Investment managers’ challenge is how to maximise the expected benefits while minimising adherence costs to these ESG-related targets. While managers will do their best, their clients must understand that their ESG goals will certainly have some performance-related effects on their portfolio returns – especially in the short term. Increased transparency should help manage client expectations and empower them to make informed decisions aligning with their values and financial objectives.
Investing in a smaller carbon footprint
Momentum Investments through its Research Hive conduct research on the impact of restricting portfolios’ carbon footprints relative to the benchmark. Carbon emissions are an important source of greenhouse gases and from an ESG perspective, investors may want to hold portfolios with a smaller carbon footprint than the benchmark. There are two main ways to do this – either exclude the most egregious emitters from the investment universe; or create portfolios using a process that down-weights them somehow.
The first approach represents a clear and simple commitment to reducing carbon emissions. It is an unambiguous signal to the offending companies, but it does result in a smaller investment universe with a strong anti-resources bias. The second method pragmatically targets a portfolio’s carbon footprint lower than the benchmark. This is more flexible and nuanced than the first approach.
These two approaches are compared in the context of large, liquid South African equities for the period 2017 to 2023 using carbon emissions data provided by MSCI Barra. Both were applied to three different factor-based investment portfolios (Value, Momentum and Quality). In each case, 10 increasingly intense carbon intensity (CI)-related targets were set and the relative performance of these more binding commitments were identified.
By carbon footprint, we mean the weighted average carbon intensity (WACI) of a portfolio (or benchmark). The WACI levels for the resources, financial and industrial super-sectors (using benchmark weights) are illustrated in Figure 1. The graph clearly illustrates that resource companies are the most intensive carbon emitters. Managing them effectively will require under-weighting this cyclical, but important, part of the market.
Eliminating carbon-intensive companies from the investment universe
The first question we asked in the research process was: what happens to your portfolios if you sequentially eliminate the top 10 most carbon-intensive companies from the investment universe? We compared the WACI for these portfolios with the WACI of the unconstrained portfolio and the average relative WACIs are summarised in Figure 2.
Excluding these shares reduces the average relative WACIs. Still, these average results hide that these reductions happen only when the excluded shares are in the top 20 when ranked by their factor exposure. In other words, the relative improvements are not consistent over time.
The next question was: what impact does optimising a portfolio to (at most) meet a benchmark relative WACI level have? Again, we evaluated 10 different levels of benchmark-relative WACI targets (decreasing by 5% each time) and constructed portfolios of at least 20 shares in a benchmark-cognisant way where we allowed a maximum of a 5% active position. These portfolios gave us the desired decrease in WACI, but more consistently this time.
The final research question was: what is the impact of these improvements in terms of the portfolios’ risk-adjusted performance? It will have an effect; either it leads to a smaller universe to invest in (ie one with fewer resources shares as they are increasingly excluded) or there is an obligation to hold a portfolio that has a lower average factor score (because we’re under-weighting the high-CI shares). The relative factor scores and risk-adjusted returns of the constrained portfolios were thus also identified. As expected, there was a negative impact on both the relative factor scores and the resulting risk-adjusted returns.
The average impact of these WACI improvements for each method over the 10 different constrained factor portfolios is summarised in Table 1. A comparison of these average WACI decreases across the various factors shows that you can constrain the carbon footprints of the portfolios and that the portfolio construction methodology does not seem to matter too much (on average).
When looking at the impacts on the risk-adjusted returns the most important point to take away from this is that in every case the decrease in risk-adjusted returns is larger than the decrease in the WACI. This decline is not an increase in risk levels, but rather a decline in the average returns from the constrained portfolios.
How to effectively manage ESG in the SA market
To summarise, our research clearly shows that while a South African equity portfolio’s WACI can be significantly improved, there are costs associated with it (lower risk-adjusted returns). The higher the desired change, the lower the risk-adjusted portfolio return no matter what method or factor considered. This is directly a result of the limited and concentrated nature of the resource-heavy equity universe that investment managers have access to in South Africa.
There are a couple of important points to note. Firstly, this exercise is limited as it is only focused on the carbon component of the E in the ESG. As an investment management firm, we believe that you can manage ESG effectively in the South African market, but as pointed out above, the nature of the South African equity market constrains the management of carbon emissions in this specific way.
Secondly, it is important to remember that this specific way of managing carbon emissions is not meant to be the only, or necessarily the most effective, way to change the desired behaviour. There is greater scope for more effective impact via active engagement with management teams by investment managers on behalf of their clients. These engagements should focus on requiring them to produce targets for ESG outcomes, develop strategies to achieve these and then hold them to account in terms of progress. As a company, we are actively engaged in these discussions through our responsible investing team.
Finally, we believe that investors need to be conscious of the potential impacts of this approach and so we remain cautious of its application in South Africa. We know that the global equity markets are a much more effective way to reduce carbon footprint by adjusting portfolio holdings as there are much greater opportunities for substituting highly CI companies.
The Research Hive was set up within Momentum Investments to provide research insights and we will continue to monitor the situation as new data arrives. ESG remains at the forefront of our investment approach. As such, we will continue to conduct these kinds of research activities. They are important to gauge what is practical for our clients as we continue to try to improve the environment and society through better portfolio management decision-making and levels of active ownership. For more information on The Research Hive, visit www.momentum.co.za. |