Rio Tinto 2023 AGM – how we are voting for net-zero accounting

Rio Tinto is committed to supporting global decarbonisation. In 2021 it set a net zero by 2050 ambition for its own operations’ (scope 1 and 2) and targets to get these emissions down by 50% by 2030. These are broadly in line with achieving a 1.5°C temperature pathway globally and should be welcomed.

The goals also make sense given Rio has a strategic opportunity, providing support for the world’s efforts to cut carbon emissions. Rio’s supply of critical minerals – such as copper, aluminium and lithium – are vital inputs into clean energy transformation. In addition, they are developing technologies that decarbonise aluminium and steel, for which there will be continued demand to support new infrastructure and economic growth.

As underscored by the new Chair Dominic Barton in Rio’s latest Annual Report to shareholders: “As we reach our 150th anniversary this year, we have an opportunity to place the energy transition at the heart of our new strategy…”.

But promises won’t be enough. The decarbonisation challenge facing Rio is enormous. It is one of the world’s most carbon intensive companies, primarily due to the use of iron ore for the production of steel, which accounts for 62% of its total emissions. A further 25% comes from processing of Rio’s bauxite and alumina. These activities delivered just over 90% of Rio’s reported underlying earnings for 2022. Decarbonisation threatens the very core of Rio’s business.

Given this, Rio’s announcement that it will spend $7.5 billion by 2030, with $1.5 billion spent between 2022 and 2024, to deliver its carbon ambitions is welcome. In the first year, however, Rio deployed less than 20% of what it had promised. At the same time, rather than emissions coming down, scope 3 emissions rose as Rio increased production of iron ore.

So, the question for Rio and its shareholders is how it can protect and enhance capital as it navigates its net-zero transition.

Key to this is having clarity as to how resilient its key assets are to write-down risk in an environment of rising carbon taxes, uncertain decarbonisation technology and shifting demand. This is at the heart of our and other investor requests over the past three years for Rio to ensure it incorporates material climate risks into its financial statements and for it to provide visibility over how it would fare in a 1.5°C-scenario.

Please see here how we voted last year and why.

Looking under the bonnet

Since we initiated our engagement, Rio has increased its disclosure on how it has considered climate-related factors in drawing up its accounts. Its latest accounts offer further improvements under a new section at the start of the financial statement titled “Impact of climate change on the Group – Strategy and approach to climate change.” Most notably, management now:

• Explicitly explains how it has reflected its promised actions to reduce operational (scope 1 and 2) emissions in its significant judgements and key estimates accounting assumptions;
• Provides the carbon price assumptions for the reference scenarios used in its core accounts sensitivity analysis; and
• Provides a Paris-aligned sensitivity analysis using its Aspirational Leadership scenario. This sensitivity foresees higher prices for copper, aluminium and high-grade iron ore (e.g. from Simandou) needed for green steel, with lower-grade iron ore (e.g. from Pilbara) seeing a fall in prices. On a net basis, Rio expects to be resilient, and may even gain.

These disclosures should be commended. They should also reassure investors that Rio is well-placed to thrive in a decarbonising world. The problem is that key points remain unclear, or raise questions, notably:

• Rio’s core accounts (and thus reported profits and capital) are predicated on global warming of well above 2°C Rio’s new base case for determining commodity prices, carbon taxes and other key inputs for accounting valuations is based on two new blended scenarios: the Competitive Leadership and Fragmented Leadership scenarios. These scenarios are associated with a world that is between 2°C to 2.5°C+ warmer than today. This new base case quietly abandons any possibility of a well-below 2°C-world that was included last year, reflecting management’s central working assumption that the Paris Agreement will fail.
• Carbon price assumptions seem low, which undermines confidence in the numbers provided and conclusions drawn. Would the results be the same if Rio used International Energy Agency (IEA) carbon prices?
  • Baseline scenarios used in core accounts: Rio assumes carbon prices will reach just $42/tCO₂e in 2030 in the two scenarios underpinning its core accounts, which raises questions:
    • How can it be that the carbon price is the same under a scenario limiting warming to 2°C versus one where warming will be above 2.5°C?
    • Why is this price assumption so far below the $75/tCO₂ used for internal planning purposes?
    • Why is Rio’s carbon price closer to an average of the IEA’s Announced Pledges scenario for developed and emerging markets, which would fall between $40 and $135?
  • 1.5°C stress test - Rio’s carbon price in the Aspirational Leadership scenario is just $59 in 2030, less than half the $140 projected in the IEA’s Net Zero by 2050 scenario (in developed markets) or $90 (for emerging markets with climate neutrality targets) – which cover the bulk of Rio’s demand.
• Ongoing failure to disclose long-term commodity price assumptions due to commercial sensitivity. Unless Rio believes that it can exert pricing power, which it suggests it cannot, it should be able to share its long-term assumptions on prices. The commercial sensitivity argument is particularly weak in the case of a 1.5°C sensitivity, given that this is a theoretical stress test. Knowing what long-term commodity prices Rio uses is essential if investors are to understand its accounting numbers, and compare Rio’s numbers to peers.
• No changes to asset lives and no impairments linked to Climate Action Plan – While it is good to know Rio has considered the Climate Action Plan in drawing up its accounts, we noted that many of the key elements of their plan do not seem to translate into changes to cashflow projections due to: 1) uncertainty in costs so cannot estimate, 2) uncertain technology so cannot estimate, 3) asset replaced planned as part of “normal lifecycle renewal”. In other words, they have been considered but not necessarily accounted for.
• 1.5°C visibility remains limited: While Rio’s conclusion that there are no material impairment risks, or changes to useful lives, or rehabilitation / restoration costs associated with its 1.5°C scenario, confidence in this scenario is undermined by 1) the lack of disclosure of what commodity price assumptions are used, and 2) the low carbon price used (see above).
• Coal exposure remains – While Rio divested of its coal businesses in 2018, they retained a contingent royalty from these assets. Due to recent coal price strength, it received cash royalties of $36 million in 2022. Overall, they have accumulated royalty receivables of $209 million at 31 December 2022. Two questions could be raised on this asset: first, are these at risk if the paying entity is unable to meet its obligations in the future as climate rules tighten? Second, is continuing to profit from coal use consistent with Rio’s net-zero commitment?

The problem with the above is two-fold.

First, if Rio’s management continues to assume commodity demand and prices associated with a world where temperatures are expected to rise well above 2°C, it will tend to deploy capital to meet these demands. It will be difficult to deliver on its own promises to support a 1.5°C-pathway.

Second, if Rio underestimates the determination of the world to drive decarbonisation, for instance through higher carbon prices, new low-carbon technologies, or even restriction on mining high-carbon minerals, Rio could be left facing stranded assets.

Vote at 2023 AGM

Taken together, we will vote as follows at Rio’s upcoming AGM:

• Financial statements: Abstain – We are abstaining, rather than voting against like last year, in recognition of improved disclosures. This includes the 1) confirmation that operational targets have been reflected in the assumptions; 2) inclusion of a 1.5°C sensitivity analysis; and 3) disclosure of carbon price assumptions, which were missing last year. We welcome the affirmation that more rapid decarbonisation would most likely be beneficial for Rio’s capital position, rather than negative. However, we continue to have concerns: 1) several of the operational costs were not included in key accounting assumptions due to the lack of certainty over costs and technologies; 2) the lack of disclosures for commodity price assumptions, particularly for the 1.5°C sensitivity analysis; and 3) the carbon tax assumptions used for both the reference scenario and the 1.5°C scenario appear to be significantly lower than the equivalent prices suggested by the IEA.
• Audit Committee Chair (Simon Henry): Abstain – we welcome ongoing improvements in disclosures on how climate change and decarbonisation is considered, but note remaining concerns (see our rationale for abstention on financial statements). We also note that Simon Henry appears to be the only member of the audit committee with accounting expertise.
• Auditor (KPMG): Against – While KPMG appears to have addressed the point regarding testing consistency with Rio’s climate action plan, they offer no comment on the reliability of the 1.5°C stress test, or on the specific carbon prices assumed by management, which we view to be too low. Moreover, despite investor requests for the disclosure of commodity price assumptions, KPMG appears to have accepted management’s position that this is not necessary without any indication of their rationale.
• Remuneration: Against - While scope 1 and 2 targets and scope 3 goals are now part of the remuneration, with an increased weight of 10% of the short-term incentive plan (STIP) component, we would like to see a 1.5°C-underpin introduced to ensure no performance-related pay is awarded for non-aligned performance.