Sustainalytics’ Carbon Risk Rating: Platypus Asset Management Live Test

Posted on April 3, 2019

Peter Brooke
Peter Brooke
Head of Quantitative Investments, Platypus Asset Management
Ying Wang
Ying Wang
Quantitative Analyst, Platypus Asset Management
Enrico Colombo
Enrico Colombo
Associate Analyst, Research Products Sustainalytics

Climate change is at the centre of public debate: from school strikes around the world to a recent landmark court ruling blocking a new coal mine in Australia on climate grounds. It is also increasingly becoming an investment risk and investors are looking to understand how this risk can affect their portfolios.

In this article, we present findings from an analysis conducted using Sustainalytics’ Carbon Risk Rating of Australian companies. The results show strong outperformance of a test portfolio built by Platypus Asset Management (Platypus) with companies assessed as low carbon risk by Sustainalytics.

Sustainalytics’ Carbon Risk Rating Framework

Measuring carbon risk can be difficult since business models and the risks they face are unique. While climate change and policy initiatives are likely to affect the economy as whole, carbon risks and impacts for a coal company are different from those faced by an insurer.

As explained in a previous article, Sustainalytics’ Carbon Risk Rating goes beyond traditional approaches of footprint measurement and screening of high emitters. Instead, the framework combines two dimensions:

  • A risk exposure assessment, based on a company’s business model, its products and operations, past performance and financial strength and;
  • An analysis of management and strategy, including governance oversight on climate risk, greenhouse gas reduction targets and renewable energy programmes.

The combination of these two dimensions results in a company’s unmanaged risk score – a forward-looking signal of how well a company is mitigating carbon risk and how well positioned it is to transition to a low-carbon economy. Investors can use this as a starting point to guide further analysis.

Platypus Analysis

Platypus is a leading Australian active qualitative and quantitative asset manager. We have been managing Australian active qualitative equity portfolios since 1999[1]. For the last decade, we have also been running systematic portfolios, where the investment decisions are based entirely on data. As a business, environmental, social and governance (ESG) issues have always been important to us and implicitly been part of how we value companies. Carbon risk, however, is something that has been more difficult to understand and incorporate. We think that Sustainalytics’ Carbon Risk Rating is a thoughtful and well-considered attempt to systematize carbon risk[2].      

Sustainalytics provided Platypus with historical carbon data going back to 2012, and we performed test analysis using our database and statistical software. We first worked through the data to remove look ahead bias, i.e. assuming that information was available at a time when it was not. Being conservative, we remove all elements in the data that are replicated year to year. This left us with a data set as detailed in the chart below.

Number of stocks with a carbon risk score

Number of stocks with a carbon risk rating score

We calculated Spearman’s rank correlation coefficient between the carbon score and returns, a method which gives a good indication of the predictability of a ranking method over time. It helps to answer the question: if we rank stocks this way, do we outperform? We restricted the comparison to stocks that had either zero for their overall risk score, or had a number, without estimating missing values. For the rest of this article, low risk means a stock has a low carbon risk score as defined by Sustainalytics.

The graph below shows the results. Month to month results can vary substantially, so we often use the 12-month average as a guide. For context, a strong predictive signal will average in the high single digits over time. Initially, low carbon risk companies did well. Over the past couple of years, this has not been the case, most likely due to the strong performance of the materials sector (which generally consists of stocks with higher carbon risk scores) during this period.

Spearman’s rank correlation coefficient for stocks with a carbon risk rating

Spearman’s rank correlation coefficient for stocks with a carbon risk rating

After calculating the rank, we split the group of stocks into thirds and looked at the performance of the lowest risk score companies compared to the highest risk score companies in the dataset. Our portfolio of low carbon risk stocks included 33 stocks in February 2012, growing to 65 in September 2018. 

Outperforming Results and Use Case

Platypus compared the return of an equally weighted low carbon risk portfolio to both the S&P/ASX 200 and to an equally weighted portfolio of higher risk companies. The results in the chart below show that the portfolio with the best third of companies would have consistently outperformed the benchmark and delivered a double-digit return over six years compared to S&P/ASX 200.

Return to an investment of $1 in a portfolio of stocks that have the lowest carbon risk, a portfolio of stocks that have the highest carbon risk, and the S&P/ASX 200 index. All dividends are reinvested, and the portfolios are rebalanced monthly. All returns are simulated using historical data.

Cumulative returns from $1 invested

The difference in investing in the lowest carbon risk companies and a portfolio of higher risk companies is the profit investors will achieve from buying low risk companies and selling high-risk companies.  Similar to the Spearman’s rank, we see that the highest risk companies have rallied over the last few years, primarily due to a rally in the materials sector over this time.

We looked more closely at the portfolio of low carbon risk companies. The portfolio generally has a higher price-to-book value than the S&P/ASX 200, and a higher price-to-earnings ratio. The return on equity of companies in the lowest risk portfolio is comparable with the market, so the portfolio (by our definitions) is neither value nor quality. Initial analysis implies that the returns are in addition to those that would be obtained through standard factors such as momentum, value, or quality. The beta of the lowest risk portfolio fluctuates around one, so investors are not taking on more market risk to achieve the returns shown above.

Sustainalytics compiled the data and Platypus performed the analysis using its database and systems. Sustainalytics did not try and pick companies that outperform, but simply considered carbon risk, therefore the outperformance is less likely to be due to over-fitting. 

We find it encouraging that investors are rewarded for buying best-in-class companies with respect to carbon risk management. Our results support the idea that there is financial value in companies’ efforts to manage and mitigate material carbon risks and report relevant related information – an approach which aligns with the recommendations of the Task Force on Climate-related Financial Disclosures. The results also demonstrate not only that integrating environmental considerations in investment decisions does not need to sacrifice return, but that investors who employ strategies considering carbon risk while constructing equity portfolios can achieve outperformance.


Important Information
The information provided in this article is general information only and current at the time of publication and does not take into account your objectives, financial situation or needs. This information is intended for recipients in Australia only. Past performance is not a reliable indicator of future performance. All returns are simulated using historical data unless otherwise explicitly stated. The content of this article is not to be reproduced without permission.

[1] See for more details, and for more information on our specific views.

[2] We are clients of Sustainalytics and are not being compensated for this article. These views are the independent views of Platypus and not Sustainalytics.