Mytilineos — Funded energy transition activities drive profitability

Mytilineos is a diversified industrial company operating in four main business areas: power generation/supply (Power & Gas division, or P&G); alumina/aluminium production (Metallurgy division); EPC for power and sustainability projects (Sustainable Engineering Solutions division, or SES); and renewables/energy storage development (RDS division). A majority of its operations are in Greece, although it has been increasing international exposure, particularly through its SES and RSD businesses. EBITDA from SES and RSD increases from a combined 8% in 2020 to 26% by 2030.

It has been investing heavily to benefit from the energy transition. It has invested c €1bn over five years (2017–21), including our estimated capex for 2021 of c €0.5bn equating to 5x depreciation, and we expect high returns (ROCE averaging 11% over our 10-year forecast period). Investments have been made in renewables, including 206MW owned renewable generation assets, mostly wind farms, and we expect a further 1.5GW of solar plants to be developed and kept (in P&G); renewable development assets (mostly solar photovoltaic, PV) of over 4.3GW at various stages of development and construction (in RSD); improving aluminium production efficiency; expanding its recycled aluminium capacity; and a high-efficiency 826MW combined cycle gas turbine (CCGT) plant, which is scheduled for commissioning during H122.

We forecast that Mytilineos increases its earnings (EBITDA) from energy transition activities from 25% in 2020 to c 60% in 2025 (and c 65% in 2030), with a significant portion of the increase over 2024–25 due to Mytilineos’s primary aluminium production qualifying as ‘low carbon’ from 2025 (see section below). This translates into a five-year EBITDA CAGR of 32% in energy transition-related activities (from c €80m in 2020 to c €320m in 2025), and a five-year group EPS CAGR of 16%. EPS CAGR is even higher over 2020–23, as by 2025 we taper down our pricing assumption for aluminium (LME price + premium) to more normalised long-term levels (we assume $2,200/t). As discussed in our decarbonisation of aluminium section below, normalising long-term aluminium prices for Mytilineos may be a conservative approach as the aluminium industry decarbonises and as customers increasingly require supply chain accountability, and may be willing to pay a premium for ‘low carbon’ aluminium. Our sensitivity analysis (Exhibit 4) shows that for every $100/t (real) increase in our long-term aluminium price assumption, our valuation increases by $2/share. Our valuation of €24.0/share is based on a normalised long-term aluminium price assumption of $2,200/t, which we keep flat (in real terms) from 2025.

Net income (after minorities) for Q321 was €38m, which is 17% above Q320 (€32m), with EPS growth over the same period even higher (we estimate at c 20%) due to the ongoing share buyback programme. EBITDA for Q321 was €86m, which is slightly (4%) below Q320 (€89m). This is mainly due to very strong prior year figures for P&G (€51m in Q320, which was an all-time high, versus €34m in Q321). Nevertheless, we are encouraged by P&G’s performance as it had the best quarter since Q320, assisted by a higher (combined) load factor at the gas thermal plants, which we estimate at 60% in Q321 compared to an estimated 39–58% over the previous three quarters (Q420 to Q221), with H121 affected by maintenance at the Korinthos Power plant.

RSD and SES both continued to show strong results in Q321, with revenue more than tripling compared to Q320 (Q321 revenue of €127m for RSD and €99m for SES). For RSD, EBITDA increased to €8m in Q321 from -€1m in Q320 driven by continued strong project delivery. The EBITDA margin for Q321 was 6%, which is slightly below our full year margin estimate of 7%. However, we expect quarterly results to be lumpy and a higher portion of full development (rather than EPC) projects in Q4 could result in a significantly higher margin. For SES, EBITDA increased to €5m from a loss of €6m, again driven by an increase in project execution. However, the EBITDA margin was just 5% for the quarter, which is lower than the previous two quarters (12% and 14% in Q121 and Q221 respectively). Similar to RSD, the delivery of a small number of large projects can make earnings for SES lumpy; we expect margins for Q421 to be notably higher.

At end Q321, net debt was €804m (as reported by management), which implies net debt to EBITDA of 2.5x. Mytilineos has a strong balance sheet with our forecast net debt to EBITDA falling below 2.0x from 2023. Furthermore, it successfully raised €500m in ‘green’ bonds in April at an interest rate of 2.25%, demonstrating investor confidence in Mytilineos’s ability to deploy capital into value accretive renewable and sustainability projects. Including the bond issue, we estimate it has financial flexibility of more than c €1.4bn (we estimate €0.5bn cash and a €0.9bn credit facility as of end Q321).

Mytilineos owns Europe’s only vertically integrated aluminium production facility (from bauxite to final aluminium products), with aluminium capacity of 250ktpa (including primary aluminium capacity of 190kt), alumina capacity of 875ktpa and nominal bauxite capacity of 570–580ktpa. It has proactively set emissions reduction targets to reduce total CO2 (scope 1 and 2) emissions in the Metallurgy division by 65% and reduce specific emissions by 75%, per tonne of aluminium produced, by 2030 (relative to 2019 emission levels). These are the most aggressive targets of any pure-play listed aluminium producer (see later in this section). Its strategy for meeting these targets includes sourcing cleaner electricity and increasing its recycling capabilities.

From 2025, we estimate that Mytilineos will qualify for S&P Global Platts’ ‘low-carbon’ classification for its entire aluminium production. This is due to an estimated 70% of electricity sourced from renewables from 2024, which results in less than four tonnes of CO2e (t CO2e) per tonne of primary aluminium produced, the threshold for a ‘low-carbon’ aluminium classification. We allow one year of ‘low-carbon’ production data to be gathered during 2024 before the aluminium is classified as ‘low carbon’ in 2025. This differs from our September report, where we made the simplification that the classification would be achieved during 2024 without significant production data being obtained.

A ‘low-carbon’ aluminium classification could potentially support above average long-term pricing premiums as customers are increasingly willing to pay a premium for low-carbon products and services, which can in turn help them achieve their own emissions reduction targets and/or make their products more attractive to environmentally conscious consumers. Furthermore, as cheap coal plants are replaced by more expensive low carbon power generation sources (not all countries have such favourable solar PV conditions as Greece), the cost curve for aluminium could steepen.

Due to its sustainability efforts, Mytilineos has achieved the Aluminium Stewardship Initiative (ASI) certification ahead of time. This requires various sustainability standards to be met, including an emission intensity for aluminium smelting of less than 8t CO2e per tonne of aluminium produced. In 2020, Mytilineos achieved 7.0t CO2e per tonne of primary aluminium produced, or 5.8t CO2e per tonne of aluminium produced (primary and secondary). We also note that Mytilineos is committed to the Science Based Targets initiative and plans to be aligned with the Task Force on Climate-Related Financial Disclosures (TCFD) recommendations in 2022. In 2021, it joined the CDP climate change initiative and also the official supporters of the TCFD initiative.

Aluminium production is one of the most carbon-intensive industries, due to the high levels of electricity required for aluminium smelting. It emits nearly 1.1bn tonnes of CO2e (t CO2e) globally (with more than 75% of this from aluminium smelting) and is not helped by China accounting for c 55% of global production. To put it into context, China used 485TWh of electricity in aluminium production in 2020, of which 80% (c 390TWh) was from coal-fired plants, and a majority of these are subcritical (low efficiency). This resulted in 667Mt of CO2e emissions from Chinese aluminium production, which is almost double the UK’s entire CO2e emissions (c 370Mt) and higher than all but the top seven CO2-emitting countries globally.

The rest of Asia, which is mostly India, produces nearly all its aluminium using coal-fired power generation, and Oceania, which is mostly Australia, produces roughly two-thirds from coal-fired power. China, India and Australia were notably absent from the coal phase-out commitments made at the recent COP26 in Glasgow. On the other hand, Europe, North America and South America (which account for c 20% of global output) produce more than 80% of their primary aluminium using low-carbon (hydro, renewable energy or nuclear) power generation. We estimate that roughly a third of global primary aluminium production uses low-carbon power sources and could potentially be classified as ‘low-carbon’ aluminium according to S&P Global Platts’ pricing classification. S&P Global Platts’ ‘low-carbon’ aluminium classification applies to primary aluminium, with maximum emissions from smelting of four tonnes of CO2e per tonne of aluminium. Excluding China, where supply chain verification is sometimes more challenging, potential ‘green’ aluminium accounts for a quarter of global production.

The aluminium industry must reduce its emissions by 77% by 2050 to meet global climate targets. This will largely be met through shifting to green electricity and assisted by increasing recycling capacity and efficiency (recycled aluminium uses c 5% of the electricity required for primary aluminium production). The International Energy Agency has called for all subcritical coal plants to be shut by 2030; this equates to the removal of c 60% of China’s aluminium production (c 30% of global supply) in less than 10 years. China has put caps on production but will need to take drastic measures to comply with global climate targets.

The challenge for the aluminium industry to decarbonise is exacerbated by only c 41% of global aluminium production residing with 11 pure-play listed aluminium producers, as well as Rio Tinto (a diversified miner), Hindalco (an aluminium and copper producer) and Mytilineos (diversified industrials). These are the companies most accountable to investors (and other stakeholders) for reducing their carbon emissions. Of the 14 companies, only eight provide emission-related data and only six of these, accounting for c 21% of global production (13.6Mt in 2020), provide carbon emissions reduction targets.

Exhibit 7 is intended as a rough guide for assessing the ambitions of those listed companies that provide emission-related data. These include many of the best-in-class aluminium producers. The remaining listed companies and other non-listed companies, comprising almost 80% of global production, typically do not provide relevant emissions-related data and emissions reduction targets. There are nuances relating to data consistency for the reported emissions intensities and corresponding emissions reduction targets. Thus, in Exhibit 7, we classify CO2e emissions intensity from aluminium smelting (normalised by primary aluminium production) according to three buckets: 1) <4t CO2e/t, which potentially equates to ‘low-carbon’ aluminium; 2) >4t CO2e/t, which indicates average emissions intensity above that required for ‘low-carbon’ aluminium; and 3) >>4t CO2e/t, which indicates average emissions intensity significantly above that required for ‘low-carbon’ aluminium and would apply to companies reliant mostly on coal-fired generation, which can equate to emissions intensities of above 15t CO2e/t. We note that even within these buckets, an averaging effect over the entire company can distort the fact that some production facilities are below 4t CO2e/t and others are above 4t CO2e/t. For example, Rio Tinto provides enough information for us to infer that its aluminium business is split between Atlantic, mostly Canada, where emission intensities at production plants are below 4t CO2e/t, and Asia-pacific, where emissions intensities are on average above 4t CO2e/t. Other large aluminium producers, with production plants split across geographies, do not provide this level of detail. Alcoa’s production facilities are based in areas including Canada, Iceland and Norway (as well as the US and Australia), so we believe some of this production is likely to be below 4t CO2e/t despite its average emissions intensity currently being above 4t CO2e/t.

In summary, Rusal is likely to be the only company whose entire aluminium production currently qualifies as ‘low-carbon’ as it uses 98% renewable energy. A significant portion of Norsk Hydro’s and Alcoa’s production also likely qualifies, as they use 70% and 78% renewable energy, respectively, although there might be potential to increase the portion as they deliver on their emission reduction targets. In particular, we note that Alcoa has production facilities in the US and Australia which run on coal. Likewise, Rio Tinto has further scope to increase its portion of ‘low-carbon’ aluminium as it delivers on its emission reduction targets, potentially reducing the carbon exposure of its Pacific aluminium business. Hindalco has significant potential to decarbonise as it is currently fully exposed to coal. However, its 25% emissions reduction target is from a 2012 base year, and it has already achieved an 18% reduction since 2012 (without moving away from coal, through measures including efficiency saving).

Mytilineos sets by far the most aggressive target of 75% reduction in emissions intensity (normalised by primary and secondary aluminium production) by 2030. Based on our review of sustainability reports for emissions reduction targets, Mytilineos appears to be the only company that is planning a material increase in secondary aluminium production. We estimate that its secondary aluminium production increases from c 20% of total production in 2020 to 40% in 2030. We estimate that Mytilineos could still achieve a 65% emissions intensity reduction based on its plans to decarbonise primary aluminium production alone (ie if secondary aluminium production were constant over the period). We estimate that Mytilineos’s emissions intensity for aluminium production falls below 4t CO2e per tonne of product in 2024 for primary aluminium production (3.2t CO2e per tonne of primary aluminium in 2024).

This enormous challenge for the supply side, which the International Aluminium Institute (IAI) estimates could cost $0.5–1.5tn, is exacerbated by strong demand for aluminium due to the energy transition. It is a lightweight material used in electric vehicles, for ‘green buildings’ and power cabling. Based on the IAI’s projections for a sub-two degrees global warming scenario, consistent with the Paris Agreement (Beyond 2°C Scenario or B2DS), demand (including recycled aluminium scrap) could increase by 80% to c 170Mt by 2050 (from 95Mt in 2018). The IAI suggests that secondary aluminium (including recycled scrap) production will increase its share from a third in 2018 (31Mt) to nearly 50% (c 80Mt) by 2050, and that up to a 40% increase in primary aluminium production is required (from 64Mt in 2018 to 90Mt in 2050), albeit produced using green electricity sources.

We use a 10-year DCF analysis (on a SOTP basis) which, we feel, considers the longer-term impact of the energy transition. We place less emphasis on peer valuation, using it only as a cross-check. Our DCF approach suggests a per-share valuation of €24.0 (rounded), which implies c 57% upside to the current share price (€15.31) and is in line with our last published valuation (in September 2021).

The key assumptions and drivers for our cash flow model are as follows:

We use peer valuation as a cross-check, rather than driving our valuation, as we believe near-term earnings multiples do not accurately reflect Mytilineos’s long-term growth prospects. In our approach, we use EV/EBITDA multiples (using current EV) applied to peers for each division. We use 2022e EBITDA given that 2021 has nearly ended.

On first impression, the peer valuation suggests €18.5/share, which is €5.5/share below our valuation of €24.0. However, we do not believe the multiples adequately reflect the competitive strengths or the growth prospects of Mytilineos’s divisions versus their peers:

We apply a 20% premium to multiples for each of these divisions/business units. Collectively, this would increase the valuation by c €5.5/share to an adjusted €24.0/share, which is consistent with our DCF valuation.

Please refer to our September outlook report for detailed financials and assumptions by division. Following the Q321 results, we made the following minor changes: