Aluminium Value Chain Is Being Redrawn: Where Returns Now Sit

Germany’s downstream aluminium exports fell 13% in the first half of 2026. Over the same period, its shipments to the US rose 228.1%. On the headline total, European rolling looks like a sector in retreat. Look at where the metal is going, and a different picture appears.

The aluminium value chain below the smelter (rolling, extrusion and foil) is being redrawn. Three forces are doing the work: US demand pulling metal across the Atlantic, China’s grip on battery materials, and India’s large but idle extrusion base. Each one changes where downstream returns are actually earned.

If you track only primary metal prices, you could misread all three. This analysis covers developments up to 10 October 2026.

You will come away with four tests for any downstream-exposed producer: tariff exposure, commissioning risk, utilisation against capacity, and margin quality against headline sales.

Why is German aluminium flowing to the US despite tariffs?

Germany consumed 1.68 Mt of flat-rolled aluminium in 2025, up from about 1.66 Mt in 2024. It still runs a trade surplus in plates, sheets, strip and foil, with exports of US$8.18 billion against imports of US$5.04 billion, a gap of about US$3.14 billion. That base is weakening.

A 10% decline figure has also circulated, but customs-based data compiled by AlCircle supports 13%. Most traditional European buyers cut purchases, while the US entered Germany’s top five destinations for the first time.

That surge of more than 47,000 tonnes came with Section 232 tariffs already in place.

The pull on German metal reflects how far the US Midwest aluminium premium has detached from global benchmarks, since a premium that wide can absorb Section 232 duties and still leave exporters better off than selling into weak European demand.

Structural reading

Industry and analyst views attributed to CRU, Wood Mackenzie and European Aluminium suggest US demand for auto body sheet, battery enclosure sheet and can stock has outgrown domestic rolling capacity. That gap keeps US conversion premiums high. These interpretations have not been independently confirmed.

On the European side, weak local demand and high power and carbon costs have narrowed margins, which pushes exporters toward better-paying US buyers.

Cyclical and tariff reading

Other analysts argue European demand is only temporarily depressed by tight monetary policy, weak construction and slower machinery orders, so flows could rebalance. The Aluminium Association backs Section 232 duties as a defence against unfairly traded metal, while European Aluminium says the duties and high premiums distort competition and encourage tariff arbitrage. US-Mexico trade talks, delayed until October, leave the duty picture unsettled.

What this tells you is that tariff headlines do not predict trade flows. Judge your exposure by asking which buyers can substitute supply and which cannot, because the second group pays the tariff.

What does North America’s rolling and recycling build-out actually deliver?

Part of the answer to that substitution question is being built in Alabama. On 7 October 2026, Novelis reported the first coil from the cold mill at its US$5 billion Bay Minette plant.

Bay Minette milestone Novelis describes it as the first fully integrated recycling and rolling plant built in the US in more than 40 years.

At full operation, the company says the site will add 600 kt a year of rolled capacity and process more than 15 billion used beverage cans annually. For US buyers relying on imported sheet, that is a meaningful future alternative.

The word “future” matters. Novelis’ guided path runs as follows:

1. First cold-mill coil, reported 7 October 2026, with remaining assets in commissioning.
2. Customer qualification and commercial shipments, guided as on track for Q2 2027.
3. Ramp-up over a further 18-24 months.

Earlier guidance was not identical. An August LinkedIn update pointed to shipments in Q1 fiscal 2028; the October release is the latest statement.

Oswego shows why one site carries so much weight. Its hot mill resumed in June 2026 after fires halted output for months, restoring supply for Ford’s F-150. During the outage Ford bought metal abroad and paid tariffs, and had estimated the disruption could cost up to US$2 billion. None of that cost evidence relates to Bay Minette.

Running the Oswego restart alongside a greenfield build concentrates execution risk, because both programmes draw on the same capital, management attention and customer commitments while the company tries to protect supply to automotive buyers.

Scrap is the other variable. A recycling-led mill needs steady scrap supply and stable scrap spreads, and heavy reliance on auto and beverage customers ties it to those cycles.

Treat a first coil as an engineering milestone. Revenue belongs to qualification and ramp-up, so that is where your earnings timeline should start.

What is a conversion premium, and why does it explain these shifts?

Both the German redirection and the Bay Minette investment rest on one idea. Four terms carry most of the analysis:

• Downstream aluminium: flat-rolled products, extrusions, foil and fabricated items made from primary metal.
• Conversion premium: the value-added margin a processor earns above the primary metal price.
• Commissioning: testing equipment and qualifying it to make specification product.
• Commercial shipments: revenue-generating deliveries to customers who have qualified the material.

The conversion premium explains direction. Where premiums are high, as in the US, metal travels toward them even with duties attached, which is why German volume moved.

Where premiums are thin, plants struggle regardless of demand. Indian extruders report exactly that, as the next section shows.

The other two terms separate a working plant from a paying one. Bay Minette is commissioning now; it is not yet making commercial shipments.

Add one more measure: utilisation, or the share of installed capacity actually running. A large plant at low utilisation spreads its fixed costs over too few tonnes.

For you, the practical point is that downstream returns depend on the premium earned and how fully plants run. The primary metal price is only the starting input.

Why is India’s extrusion sector running at 30% utilisation despite demand growth?

India has the capacity. It does not have the output.

Utilisation check India’s extruders are running at about 30% of installed capacity, according to AlCircle’s September analysis.

Demand is growing about 8%, and imports still exceed local output. Growth is not reaching domestic plants.

The duty structure helps explain why. Primary aluminium carries a 7.5% basic customs duty, about 8.25% effective, while many finished downstream products enter from ASEAN and the UAE at zero or near-zero duty under trade agreements. Domestic extruders pay more for their raw material than importers pay for the finished good.

The industry body ALEMAI has asked for zero duty on ingots and billets and higher duties on finished products. Primary producers counter that the duty protects smelters and fiscal revenue, and trade commitments limit how far finished-goods duties can rise. Add working-capital strain, and many extruders simply choose not to run.

Specialised capacity is the contrast. Sharvaya Metals commissioned a 10-inch press on 5 October 2026, aiming at defence, aerospace, railways and EVs.

Read India as a warning. Capacity built ahead of demand and exposed to duty-advantaged imports does not turn growth into returns, so check utilisation and import competition before you value any capacity figure.

For readers wanting the full picture behind the utilisation numbers, our deep-dive into India’s extrusion output collapse traces why production fell so sharply against rising demand.

How exposed is Asia’s battery foil supply to China?

The import data

South Korea imported 57,261.18 tonnes of aluminium foil in the year to July 2026, up 23.94% from 46,201.68 tonnes, a five-year high. Battery and mobility demand is feeding it. Lotte Aluminium estimates each GWh of battery capacity needs about 700 tonnes of electric foil, and Korea has lined up KRW 400 billion in government R&D over five years plus KRW 8 trillion in industry commitments for 2026-2030.

Vehicles are using more of the metal too. Jaguar’s Type 01 is 73% aluminium, and Türkiye’s TÜRASAŞ plans 14 aluminium-bodied high-speed trains across 2027-2028.

Then comes the source. China supplied 52,751.81 tonnes, or 92.12% of Korea’s imports, and about 96% of the 11,059.5-tonne increase.

Chinese foil exports reached 802,000 tonnes in January to July 2026, with some forecasts pointing to roughly 1.3 Mt for the year. Full-year data are not yet available.

What could go wrong

The supply channels most exposed to disruption include:

• Chinese export controls
• Power rationing affecting smelters and mills
• Trade tensions between China and buyer countries
• Currency and logistics shocks

Some analysts warn that China’s earlier tightening of graphite and rare earth exports, which drove price spikes and sourcing reviews, could repeat in foil. That warning has not been independently confirmed and should be read as a scenario.

Asian battery growth and Chinese supply risk are one story. If you hold exposure to either, you hold the other, along with the qualification hurdles any non-Chinese supplier must clear.

What should mining and energy investors take from the downstream shifts?

The four regions point to one lesson: tonnage is where the analysis begins, not where it ends. Golkonda Aluminium Extrusions illustrates why. For the quarter to 30 September 2026, it reported sales of INR 13.99 billion (US$144.48 million) yet a net loss of INR 0.321 million, against net income of INR 0.471 million a year earlier.

Those figures come from company results reported via AlCircle without independent confirmation, and the year-ago sales comparison appears inconsistent in scale. Treat them as an illustration, not a benchmark.

The mechanism is common across the sector. Extruders cut prices to keep lines running, absorb higher energy and labour costs, and extend credit terms, all of which erode EBITDA (earnings before interest, tax, depreciation and amortisation) per tonne.

Apply the tests in this order:

1. Confirm the plant actually runs (utilisation).
2. Confirm when it earns (commissioning and qualification).
3. Confirm who buys and under what duties (tariff and customer concentration).
4. Confirm what each tonne earns (margin quality, contract terms, recycled-input share).

For rolling assets, the questions sharpen around long-term contract quality, recycled-input share and single-site exposure. Oswego showed how one outage can become an earnings shock.

Reading the redrawn map: what to watch from here

Value is moving toward places where premiums, policy support and qualified demand line up, and away from idle capacity and single-source supply.

Three markers will test that reading. Watch Bay Minette’s qualification progress toward Q2 2027, the outcome of US-Mexico trade talks and any Section 232 decisions, and any change to India’s duty settings or fresh utilisation data.

The US-Mexico talks and any Section 232 decisions matter because tariff policy is increasingly tied to US smelter investment, so duty changes can reshape both import costs and the long-run domestic supply base.

Each marker answers one of your four tests. Use them to decide which downstream exposures earn a premium and which merely report volume.

Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors. Forward-looking statements are speculative and subject to change based on market developments and company performance.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions.