US Alumina Imports Rise 2.7%, but a 25% Q2 Drop Tells the Real Story

A 2.7% rise in half-year alumina imports looks like steady, unremarkable demand. It is the least useful number in the dataset. Underneath it sit a 32% surge in the first quarter and a 25% fall in the second, and those two quarters point in opposite directions.

Alumina is the refined powder that primary aluminium smelters turn into metal. The US is restarting idled smelters while still depending heavily on imported feedstock, so US alumina imports offer an early read on whether that revival has supply behind it.

According to AlCircle, which published the figures on 7 October 2026, the US imported 712,000 tonnes in H1 2026 against 693,000 tonnes in H1 2025. These totals could not be independently confirmed in Census or United States Geological Survey (USGS) data visible online, so treat them as a single-source picture.

This piece shows which suppliers drove the swings, what the quarterly volatility can and cannot tell you, and which feedstock risks matter if you hold Mining & Energy exposure.

Why does a 2.7% half-year gain hide a sharp Q2 reversal?

Taken over three years, the trend looks healthy. Imports climbed from 495,000 t in H1 2024 to 693,000 t in H1 2025, then edged up to 712,000 t this year.

Split the half into quarters and the picture changes.

Q2 2026 volumes of 266,000 t came in well below the 356,000 t of Q2 2025. The entire half-year gain rests on the first quarter.

The quarterly split Q1 2026: +32%. Q2 2026: -25%. The headline gain exists only because Q1 was strong enough to absorb the Q2 slump.

No analyst or institution has explained the swing. One plausible reading, offered here as interpretation rather than sourced fact, is that buyers restocked in Q1 ahead of smelter restarts, then pulled back once inventories were built. Short-term curtailments or some domestic substitution could also have weighed on Q2. Reuters reported on 9 July 2026 that an “aluminium supply shock” had tightened global supply, which fits a restocking story without proving it.

Alumina price trends add another layer to the import picture, since feedstock costs shape smelter margins and can influence how aggressively operators restock after a strong first quarter.

What this tells you is that a single half-year figure is a poor guide to feedstock demand. Momentum entering H2 2026 looks weaker than the headline suggests.

Which suppliers drove the swings: Brazil’s grip, China’s climb and Jamaica’s collapse

Brazil: the dominant supplier

Brazil supplied 568,000 t of the 712,000 t total, roughly four tonnes in every five. That is up about 5% on 542,000 t in H1 2025 and well above 430,000 t in H1 2024.

Its quarterly pattern mirrors the overall market almost exactly. Brazilian shipments rose 27% to 342,000 t in Q1, then fell 17% to 226,000 t in Q2, marginally below Q2 2024.

When one country moves this much tonnage, the total import line is largely a Brazil line. Based on general industry knowledge rather than this dataset, Brazil’s position likely reflects world-scale refineries, integrated bauxite supply and established shipping routes to North America.

China posted the strongest half-year growth among listed suppliers, but its gains came early. Q1 rose to 19,000 t from 16,000 t, while Q2 held flat at 7,700 t against 7,800 t. Canada and Germany delivered small, steady volumes.

Double-digit growth from these suppliers looks impressive in percentage terms. In tonnage, it does not diversify the market in any meaningful way.

Jamaica: a 76% drop with no clear explanation

Jamaica produced the sharpest move. Shipments fell from 58,000 t to 14,000 t, with zero imports in Q1 and a resumption of 14,000 t in Q2.

On Century Aluminum‘s Q1 earnings call on 8 May 2026, management said a new steam turbine at the Jamalco refinery was expected to start production that month, cutting costs and reliance on grid power. That confirms Jamalco was operating and investing. It says nothing about export volumes.

No source explains the decline. Higher energy costs and a history of outages are general industry factors, not proven causes here, so the drop reads more as a signal of supply-side fragility than of weaker US demand.

Jamaican alumina export swings have a history of moving the US import line, and the 2026 collapse follows a period when shipments from the island had roughly doubled, which makes the current drop harder to read as a simple demand signal.

What is alumina and why do smelter restarts matter for import demand?

How alumina feeds aluminium

Alumina is refined aluminium oxide, traded mainly as calcined alumina under customs code HS 281820. It sits in the middle of a three-step chain: miners dig bauxite ore, refineries process it into alumina, and smelters use electricity to convert alumina into aluminium metal.

A smelter cannot run without a steady alumina supply. When smelters restart, feedstock demand follows.

Western smelter restarts reflect decades of capacity loss now being partly reversed, and the 2026 supply shock has made even small potline returns commercially attractive for operators with secure power.

That matters because the US has very little smelting left. USGS data show primary output of about 660,000 t in 2025, against roughly 4.4 million t of imported crude and semi-fabricated aluminium.

A long decline According to SAFE’s Center for Strategic Industrial Materials, US primary aluminium output fell from 4.9 million t/year in 2000 to under 710,000 t/year in 2024.

The restarts lifting US demand

Two projects anchor the current revival:

• Century Aluminum, Mount Holly (South Carolina): a restart of more than 50,000 t/year, backed by a $50 million investment and an extended power deal with Santee Cooper. Reports point to near-full production by 30 June 2026, taking capacity to about 220,000-230,000 t/year and lifting US primary output by about 10%.
• Magnitude 7 Metals, New Madrid (Missouri): one 75,000 t/year potline planned by end-2026, according to Reuters, with full capacity of 263,000 t/year possible over time, per Bloomberg.

No source quantifies how much extra alumina these restarts require. The link to import demand is directional only.

What you should take from this is that feedstock needs rise gradually, in step with power availability. Read import volumes against smelter operating rates, not as a stand-alone indicator.

How exposed are US smelters to feedstock and power risks?

The restarts are real. They are also small against the gap they are meant to close.

The Aluminum Association’s 2026 roadmap says slightly less than half of installed US primary capacity is operating, and that restarts at four existing smelters could replace only 15-20% of current primary imports. SAFE estimates that even a fully running New Madrid would take US output to just over 1 million t/year.

The structural gap SAFE projects a 3 million tonne US aluminium import deficit by 2029, even with New Madrid at full capacity.

Tariffs do not appear to be the deciding factor. Century credited Section 232 aluminium tariffs with helping the Mount Holly restart, while Steel Market Update argued on 6 February 2026 that the Santee Cooper power agreement was the main driver. Reuters reported on 18 February 2026 that another US smelter shut despite tariff protection, and Fastmarkets wrote in October 2025 that smelter expansion depends on energy costs and long-term power contracts.

Existing tariff commentary also covers aluminium metal, not alumina or bauxite. Feedstock enjoys no comparable policy shield.

The counter-argument deserves weight. Mount Holly and New Madrid show domestic capacity can rebound, and every restart trims import reliance. Both SAFE and the Aluminum Association agree, however, that a large import requirement remains even at full utilisation.

Concentration compounds the exposure. With Brazil supplying about 80% of H1 2026 volume, a geopolitical or shipping disruption in one country would hit most of the feed.

Three watch-points matter most for your exposure:

1. Power costs and contracts: new or extended long-term power deals are the clearest signal a restart will hold.
2. Brazil concentration: any change in Brazilian shipments moves the total import line almost one for one.
3. Jamaican refinery reliability: whether Jamalco’s turbine stabilises output enough to restore volumes.

Restarts reduce feedstock risk. They do not remove it.

What the quarterly swings tell you, and what they cannot

The H1 2026 data carry three findings. The modest headline gain masks a Q2 slump, Brazil effectively sets the direction of the total, and smelter restarts leave the US structurally dependent on imported feedstock.

The limits are just as clear. The figures rest on a single source, and nothing in the public record explains the Q1 surge, the Q2 slump or Jamaica’s collapse.

The signals worth tracking are Q3 import data, smelter operating rates at Mount Holly and New Madrid, Jamalco’s output after the turbine start-up, and fresh power-contract announcements. Together, they will confirm whether Q2 was a pause or the start of a softer trend.

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. Past performance does not guarantee future results. Financial projections are subject to market conditions and various risk factors.