As discussed in Section 5, various collaborative initiatives encourage investors to set targets to reduce emissions in their investment portfolios. While such initiatives are a welcome and necessary step, reducing carbon exposure within a listed equities portfolio may not result in immediate emissions reductions in the real economy. In addition, such initiatives do not guarantee, and in some cases work against, forceful and exhaustive engagement with carbon intensive companies.
By reducing carbon exposure within a listed equities portfolio, investors aim to reduce their exposure to transition risk faced by the most carbon intensive companies. As the world shifts away from fossil fuels, companies responsible for producing or burning coal, oil and gas will become less valuable. However, transition risk is only one form of climate risk, and investors focused exclusively on reducing carbon exposure may overlook companies exposed to the fossil fuel supply chain (which are not necessarily carbon intensive), or companies exposed to the physical risks from climate change.
Further, reducing portfolio carbon exposure does not grapple directly with the systemic risks of climate related upheaval. To put it another way, not even a carbon free portfolio will guarantee strong returns in a scenario with extreme or runaway climate change. This is because reducing carbon exposure does not necessarily equate to reducing emissions in the real economy. This is acknowledged by the Net-Zero Asset Owners Alliance, which notes that the “decarbonisation of portfolios could be easily achieved by selling carbon intensive investments. However, it is highly questionable if such actions alone would have a positive impact on the real economy”. Counter to this argument, divestment advocates have long contended that, at scale, divestment of fossil fuel companies will not only reduce the share price of those companies, but in the long run, increase their cost of capital, putting pressure in a material sense on projects on the margins of viability. The DivestInvest initiative encourages individuals and organisations to divest from fossil fuel companies and invest in climate solutions. This reinvestment of capital is intended to complete the act of divestment from fossil fuels, by achieving the aforementioned goals and encouraging those companies dedicated to reducing emissions.
According to MSCI, the Australian share market is one of the most carbon intensive indexes in developed markets. Within the benchmark index S&P ASX/200, much of the carbon intensity is driven by a relatively small number of companies. As shown below, more than 80% of the Weighted Average Carbon Intensity of the S&P ASX/200 comes from just three sectors: Materials, Utilities and Energy.
The relatively small size and concentration of the S&P/ASX200 provides limited opportunity to reduce carbon intensity through stock selection within the same sector. For example, as at 30 September 2020, there were only four companies in the S&P/ASX200 Utilities sector, thus making it difficult for investors to switch out of the most emissions intensive company in the entire index, AGL Energy. This is less of an issue for international equities portfolios, as the larger number of companies and geographical representation in global benchmarks, such as the MSCI All Countries World Index (ACWI), provide greater opportunity to reduce carbon intensity through stock selection.
Australian superannuation funds with approximately similar exposure to Australian and international equities can significantly reduce their overall carbon intensity by shifting to a low carbon benchmark in the passive component of their international equities portfolio. Reducing carbon intensity in Australian equities portfolios is likely to prove more difficult.
Aware Super, upon announcing its suite of climate targets in July 2020, set a target of “a minimum 30% reduction in emissions in [its] listed equities portfolio by 2023, which will also incorporate the introduction of a new low-carbon index”. This was clarified in its announcement of a 40 per cent reduction in the carbon footprint of its equities holdings “virtually overnight” in November 2020, through the divestment of 60 companies globally, including nine in Australia. Aware Super declined to name the companies it divested.
Superannuation funds have a responsibility to members to be transparent about what reducing carbon exposure means in practice, and what it may or may not mean for emissions reductions in the real economy and their engagement with listed companies. Where divestment is likely to reduce exposure, but have limited or no impact on emissions in the real economy and potentially put engagement outcomes further out of reach, funds should speak clearly about this.
While carbon exposure metrics are a useful tool for measuring and managing the transition risks from climate change in a listed equities portfolio, several issues arise in attaching targets to such metrics, primarily relating to data reliability and sufficiency.
ACCR assessed the appropriateness of using carbon exposure metrics to assess the transition risk faced by the 25 largest emitters in the S&P/ASX200 index. The 25 largest emitting companies in the S&P/ASX200 index constitute 26.6% of the index weight, but 72.8% of the Weighted Average Carbon Intensity (metric tons CO2e/USD 1 million revenues).
We found that the carbon exposure metrics defined in Section 6 are insufficient for managing the breadth of transition risks faced by Australian listed companies.
We recommend that investors ask the following questions before setting targets based on carbon exposure metrics:
Is the emissions data reliable?
What are the company’s Scope 3 emissions?
What are the company’s short- and medium-term plans?
Has the company achieved real world emissions reductions?
7.1 Is the emissions data reliable?
In Australia, listed companies are not legally required to disclose their greenhouse gas emissions in their annual reports. Corporate groups emitting more than 50,000 tonnes of CO2-e emissions (Scope 1 and Scope 2) are required to report to the Clean Energy Regulator, and this data is then disclosed to the general public in the March following the end of the financial year (30 June). However, due to complex corporate structures, many listed companies have multiple entities in the Clean Energy Regulator dataset, particularly joint ventures or where a company does not have operational control. Rio Tinto, for example, has at least five different entities in the Clean Energy Regulator dataset, including joint ventures. For this reason, company disclosures remain the single most important source of accurate and timely emissions data.
According to the Australian Council of Superannuation Investors (ACSI), just 60% of S&P/ASX200 companies disclose all their operational emissions (Scope 1 and 2).
ACCR found that the 25 largest emitters in the S&P/ASX200 all disclose their operational emissions (Table 4). However, several companies do not disclose emissions data for the same period as their financial reporting, including Alumina, BlueScope Steel and Santos. Best practice disclosure of emissions should be consistent with financial reporting, and include a breakdown by commodity or business, allowing for a better understanding of the progress of each business unit or facility over time. Such disclosure is currently the exception rather than the norm.
In the absence of company disclosures, index providers such as S&P use estimates, typically based on the sector in which a company operates. This lack of accurate and timely data from all companies within an index must be considered when investors are deciding to reduce their exposure to specific companies or sectors.
This issue would be best addressed by the introduction of a legal requirement for companies to provide accurate, timely and audited emissions disclosure, and accompanied by fines for non-compliance. In the absence of regulatory reform, investors should thoroughly interrogate company level data and seek to fill gaps.
7.2 What are the company’s Scope 3 emissions?
Arguably, the most significant problem with carbon exposure metrics is the exclusion of Scope 3 emissions (value chain emissions). The carbon exposure metrics defined in Section 6 do not include Scope 3 emissions, and currently there is no standard metric applied to listed companies that captures Scope 3 emissions. In 2019, ACSI found that just 57 out of the S&P/ASX200 companies, or 28.5%, disclosed Scope 3 emissions. Disclosure of Scope 3 emissions by the largest emitters in the S&P/ASX200 index is better than the broader index, which is likely due to greater investor scrutiny around climate risk disclosure at those companies.
ACCR found that 20 of the 25 largest emitters in the S&P/ASX200 disclose some Scope 3 emissions data (Table 4). 13 companies have significantly higher Scope 3 emissions than their operational emissions. Whitehaven Coal does not disclose its Scope 3 emissions, but they are likely to be much greater than its operational emissions.
For companies involved in the fossil fuel supply chain, Scope 3 emissions typically dwarf their operational emissions. For example, Santos and Woodside Petroleum’s Scope 3 emissions are approximately four times the size of their operational emissions, while BHP’s Scope 3 emissions are more than 35 times the size of its operational emissions.
Investors that disregard Scope 3 emissions may remain exposed to significant transition risk. Some companies taking active steps to reduce their operational emissions are also increasing fossil fuel production. Recently, Santos and Woodside Petroleum committed to reduce their operational emissions by 26-30% and 30% by 2030, respectively. Both companies plan to achieve this through a combination of carbon offsets, energy efficiency and carbon capture and storage. However, both companies are also planning to significantly increase gas production over the same timeframe, and have avoided setting Scope 3 targets.
It is also worth noting the issue of ‘financed emissions,’ that is, the carbon exposure of companies in the financial services sector. While an analysis of financed emissions profiles of ASX-listed financial services companies is beyond the scope of this report, asset managers, banks and insurers are likely to be exposed to carbon emissions an order of magnitude greater than their own operational emissions. These emissions are also not captured by carbon exposure metrics.
7.3 What are the company’s short- and medium-term plans?
Carbon exposure metrics are, by their nature, based on historical data, and do not necessarily reflect forward-facing business strategies. There may be companies with ambitious short-to medium-term emissions reduction commitments with relatively poor carbon exposure metrics. By contrast, a falling trend in historical emissions, particularly as a result of COVID-19, may have little relation to the future climate impact of a fossil fuel company with aggressive expansion plans.
Two pillars of the TCFD framework—scenario analysis, and metrics and targets—were designed to give investors clarity around a company’s resilience to the various aspects of climate risk. Neither of these forward-looking disclosure categories are captured by carbon exposure metrics.
ACSI found that just 37% of S&P/ASX200 companies have set emissions reduction targets, “with a noticeable shortage of medium and long-term aims”. Furthermore, ACSI identified just 28 companies with medium term targets (to 2026-2039) and 13 companies with long term targets (to 2040 and beyond).
ACCR’s analysis of the 25 largest emitters in the S&P/ASX200 found that:
Nine companies have short-term targets to reduce emissions by 2025 or sooner. Five of these are “intensity” targets (reductions in emissions per unit of production), while three of the nine companies have set short-term targets that require absolute emission reductions. One company, Wesfarmers, has separate short-term targets for its subsidiaries which includes both intensity and absolute targets.
Ten companies have set medium-term emissions reduction targets by 2030. Three of these are intensity targets and six require absolute emission reductions. Additionally, Rio Tinto has set both an intensity and absolute medium-term emissions target for 2030.
Ten companies have set net-zero emissions targets by 2050, and one company, Woodside, has an “aspirational” target of net-zero by 2050. None of these net-zero commitments include Scope 3 emissions.
Five companies have set targets related to their Scope 3 emissions: Amcor, BHP, Boral, Origin Energy and Woolworths. Two of those companies are involved in fossil fuel extraction—BHP and Origin Energy—but neither has committed to reducing fossil fuel production.
Two companies have emissions reduction targets certified by the Science-Based Targets Initiative (SBTi): Origin Energy’s target is aligned to a 2°C pathway and Woolworths’ target is aligned to a 1.5°C pathway. BHP is currently seeking SBTi certification.
Four companies have not set any emissions reduction targets at all: APA Group, Ausnet Services, Viva Energy and Whitehaven Coal.
There has been a spate of new emissions reduction targets announced by Australian listed companies in 2020, particularly commitments to net-zero by 2050. Investors must ensure that these long-term targets are complemented by ambitious short- and medium-term targets.
7.4 Has the company achieved real world emissions reductions?
In recent years, several Australian listed companies have reported declines in operational emissions that have come about through the divestment of assets or changes to their supply chains. Crucially, however, these have not resulted in emissions reductions in the real economy. We explore some of these examples below, with a view to informing decision-making around how emissions reductions are achieved in listed equities portfolios.
Corporate asset-level divestment
Rio Tinto sold its NSW coal mines to Yancoal in 2017, and its Queensland coal mines to Glencore, EMR Capital and Adaro Energy in 2018. In its 2019 climate change report, Rio Tinto declared that it had “reduced the absolute emissions from [its] managed operations by 46%” since 2008. Excluding the divestment of its coal assets, Rio Tinto only reduced its operational emissions by 18% between 2008 and 2019. Crucially, each of the operating coal mines sold by Rio Tinto continue to produce coal. In the case of the Valeria coal mine, Glencore is seeking approval from the Queensland government to commence production by 2026.
Similarly, BHP Group spun off numerous emissions-intensive assets into South32 in 2015, then divested its onshore US oil and gas assets in 2018. Largely because of these divestments, BHP’s absolute emissions declined from 45 million tonnes CO2e in FY14 to 15.8 million tonnes CO2e in FY20. The assets divested by BHP in 2015 and 2018 continue to produce carbon emissions today.
Following sustained pressure from institutional investors, in August 2020, BHP CEO Mike Henry announced that BHP would divest its remaining thermal coal mines—Cerrejón in Colombia and Mt Arthur in NSW (Australia)—within two years. BHP also committed to divesting its 50% stake in the BHP-Mitsui metallurgical coal joint venture, and its 50% stake in oil and gas fields off the coast of Victoria. While these additional divestments will likely improve BHP’s carbon exposure metrics, the carbon emissions do not simply disappear.
A different, if tentative, approach has recently been advanced by outgoing Glencore CEO Ivan Glasenberg, who has criticised the push to divest coal assets, suggesting that they could end up in “the hands of other players who have no intention of reducing Scope 3 emissions and if anything gives them a free hand to start producing more”.  At face value, these comments suggest that Glencore may be prepared to begin winding up fossil fuel producing assets, rather than divest them, to reduce its Scope 3 emissions. If matched with action in the near term, this would be a welcome development.
Companies may also be motivated to divest fossil fuel assets as they near end-of-life to avoid the costs and complexities of a “just transition” for workers, as well as significant rehabilitation costs. Long term investors should be sensitive to the need to avoid scenarios where scrutiny of a company’s Scope 3 profile results in abrupt divestment of an asset to a buyer of last resort that is disinterested in emissions reduction, workforce transition, responsible decommissioning or environmental remediation. Recent months provide a case in point: the Australian Government has been forced to cover the cost of decommissioning for the Northern Endeavour oil platform, which was operated by Woodside between 1999 and 2016. Woodside paid $24 million to Northern Oil and Gas Australia (NOGA) to take over the site, but NOGA subsequently entered voluntary administration in late 2019.
IIGCC’s Net Zero Investment Framework (Section 5) recommends investors vote against mergers and acquisitions (M&A) where the post M&A company cannot meet the same expectations of emissions reductions.
Changes to supply chains
In 2012, Boral announced that it would cease manufacturing clinker—a key ingredient in cement production—at its Blue Circle Southern Cement plant in Victoria. Then CEO, Mike Kane, said at the time that the combination of a high Australian dollar, low shipping costs and the increased cost of local manufacturing made importing clinker relatively cheaper than manufacturing it locally. In FY2020, Boral’s clinker manufacturing operations in Australia accounted for 45% of its total Scope 1 and 2 emissions, producing 70% of Boral’s clinker, while the remaining 30% was imported. Since FY2005, Boral has reduced its Scope 1 and 2 emissions in Australia by approximately 44%. Boral admits that it “achieved this largely by realigning [its] portfolio away from emissions-intensive businesses and reducing clinker manufacturing in Australia”.
In FY2020, Boral’s Scope 3 emissions were 3.1 million tonnes CO2e and its operational emissions were 2.2 million tonnes. While Boral’s disclosure of its operational emissions is quite comprehensive (providing seven years’ worth of data), the company only disclosed its Scope 3 emissions for the first time in its FY2019 reporting. This lack of data prevents any analysis of the impact of offshoring clinker production.
Closure of old facilities
At the end of March 2017, French company ENGIE closed its majority-owned Hazelwood coal-fired power station in Victoria, after just five months’ notice. ENGIE cited lower electricity prices and a surplus of electricity supply as the reasons for its closure. While ENGIE was not a member of the S&P/ASX200 index, the closure of Hazelwood had a significant impact on the index’s carbon intensity.
In FY2016, Hazelwood produced more than 10.3 GWh of electricity and 14.6 million tonnes of CO2e emissions. Hazelwood’s relatively abrupt departure from the National Electricity Market (NEM) required other carbon intensive generators to increase their output, including AGL Energy and Origin Energy. So, while the closure of a brown coal-fired power station was a positive step for climate action, the consequence was increased output and therefore emissions from at least two ASX-listed companies.
Three years after the closure of Hazelwood, Origin Energy estimated that the closure resulted in its Scope 1 equity emissions increasing by more than 1 million tonnes CO2e in FY2018 and FY2019, as it “supported the market with [its] thermal generation fleet” to “ensure security of supply.”
Changes in market dynamics are not uncommon, and there may be future scenarios where a large power station or industrial facility undergoes a planned or unplanned temporary closure that results in competitors increasing production. Such incidents, while temporary, may affect the carbon exposure metrics of multiple market participants.
ACCR’s climate program aims to accelerate Australia’s transition to a low carbon economy in line with the Paris Agreement. We engage with ASX-listed companies on their climate risk disclosure and the need to set emissions reduction targets consistent with the Paris Agreement, and we also push for reviews by ASX-listed companies of their industry associations’ climate policy advocacy.