London Metal Exchange (LME) cash aluminium reaches $3,600 per tonne by December 31, 2026 in the base case, $3,950 in the bull case and $2,900 in the bear case, driven not by production cost but by the collapse of freely deliverable exchange inventory to its lowest level since records began in January 1998.

LME cash aluminium settled at $3,327.50 per tonne on August 10, 2026, with exchange stocks at 254,900 tonnes (LME official data via Westmetall). That inventory figure is 42.7% below late January and the lowest in the exchange’s 28-year record. The base case argues that the market repriced the Gulf war premium out of aluminium between June and July while the inventory deterioration continued unpaid for. The thesis breaks if any one of four signals fires, listed in the disconfirmation section.

Key Levels:

• Aluminium (LME cash): $3,327.50/t settlement, August 10, 2026 — LME official data via Westmetall; three-month $3,320.50/t
• Base case target: $3,600/t by December 31, 2026 — re-test of the June 2026 four-year-high zone as warrant scarcity is repriced
• Bull case target: $3,950/t — if Gulf smelter restarts slip past year-end and Strait of Hormuz transit re-tightens
• Bear case target: $2,900/t — if the Indonesian ramp lands on schedule and LME stocks rebuild above 400,000 tonnes
• Major support: $3,085/t — the four-month low set in July 2026 (Trading Economics, August 10, 2026)
• Major resistance: $3,750–$3,800/t — the early-June 2026 four-year high, still short of the March 2022 all-time high
• Invalidation level: two consecutive weekly closes below $3,085/t on LME cash

Methodology and what this call is built on

Price data is LME official cash settlement and three-month closing prices sourced from Westmetall’s daily republication of exchange data, cross-checked against Trading Economics futures quotes, both retrieved on August 11, 2026. The spot anchor is the August 10, 2026 settlement, not a weekly or monthly average. Inventory data is LME warehouse stock and on-warrant tonnage. Production data comes from Shanghai Metals Market (SMM) monthly analyses and company disclosures. Alumina and bauxite references are S&P Global Platts assessments.

Three caveats apply. LME stocks capture a shrinking share of global inventory as off-warrant metal grows, so the exchange number overstates absolute scarcity while still signalling directional stress. Origin-composition data for LME warrants is published with a lag, and the most recent robust reading is May 2026. And the forecast horizon spans a Gulf restart schedule being revised roughly monthly.

The data: stocks fell 42.7% while the price fell 12%

The central anomaly in aluminium is a divergence. Prices peaked near $3,800 per tonne in early June 2026, a four-year high, corrected roughly 16% to a four-month low around $3,085 in July, then recovered to $3,327.50 on August 10. Over the same window exchange inventory went one direction only.

Variable	Latest	Prior reading	Change	Source and date
LME cash settlement	$3,327.50/t	$3,015/t (Jan 2)	+10.4% YTD	LME via Westmetall, Aug 10, 2026
LME total stocks	254,900 t	~445,000 t (late Jan)	−42.7%	LME via Westmetall, Aug 10, 2026
LME on-warrant stocks	245,350 t	271,450 t (Aug 2022 record)	−9.6%	Reuters, Jul 24, 2026
Russian share of available stocks	93%	72% (April 2026)	+21 pp	LME origin data, May 2026
Output outside China, July y/y	−6.7%	—	−6.7%	SMM, Jul 31, 2026
China operating capacity	46.29 mt	45.00 mt policy cap	+2.9% over cap	SMM, end-July 2026
Alumina FOB Australia (Platts)	~$365/t	~$700/t (Q4 2024 peak)	−47.9%	S&P Global Platts, mid-July 2026
US Midwest premium	$2,182/t	~$2,000/t (early Jan)	+9.1%	Platts assessment, Feb 2026 record

Sources: LME official data via Westmetall (August 10, 2026); Reuters (July 24, 2026); Shanghai Metals Market (July 31, 2026); S&P Global Platts. Time window: January 2, 2026 to August 10, 2026.

LME aluminium inventories are the lowest since the exchange’s records began in January 1998, but the headline tonnage understates the tightness. Total stocks settled at 254,900 tonnes on August 10, 2026, against roughly 445,000 tonnes in late January — a 42.7% drawdown in seven months. On-warrant metal, the portion actually deliverable against a contract, stood at 245,350 tonnes on July 24, below the previous record low of 271,450 tonnes set in August 2022. The composition matters more than the total. Russian-origin metal accounted for about 93% of available LME stocks in May 2026, up from 72% in April, and Russian plus Indian units together made up roughly 96% of warrant-eligible inventory. That leaves roughly 4% — under 10,000 tonnes — from Middle Eastern, European and other origins, inside a total that Reuters calculated as less than one day of global consumption.

“The aluminium market is so much larger today than it was 25 or so years ago, so LME stocks relative to the market size are very, very low today.”

— Duncan Hobbs, Research Director, Concord Resources (Mining Weekly / Reuters, July 27, 2026)

The mechanism: this is a warrant problem, not a cost-curve problem

The intuitive case for higher aluminium is a cost-floor argument: power is expensive, smelting is power-hungry, so the marginal cost of production sets a floor under the price. In August 2026 that argument runs backwards. Aluminium’s largest single input cost is alumina, consumed at roughly 1.93 tonnes per tonne of metal, and the Platts FOB Australia alumina assessment sat near $365 per tonne in mid-July 2026 against a Q4 2024 peak above $700. Guinean bauxite fell to roughly $32–$38 per tonne FOB by late April 2026, the weakest since March 2022. Smelter margins are therefore unusually wide, which is an argument for more output, not less. The bullish case for aluminium does not rest on the cost curve. It rests on the fact that the metal which exists is in the wrong places, in the wrong ownership, or behind the wrong tariff wall.

Three walls are doing the work. Sanctions and buyer policy make Russian units, now the overwhelming majority of LME warrants, unusable for a large share of Western consumers. The US Section 232 tariff on primary aluminium, raised to 50% in mid-2025, has pushed the Midwest premium to a record $2,182 per tonne, siphoning non-Russian units across the Atlantic rather than into exchange sheds. And China, which produced 22.34 million tonnes in the first half of 2026 against a 45-million-tonne policy cap, is exporting fewer castable ingots: SMM put July casting-ingot output down 15.1% year on year as smelters diverted a record 78.3% of production to liquid metal sold straight to domestic processors. Metal that never solidifies into an ingot can never become a warrant.

The steelman for the other side is serious: supply is coming. SMM reported overseas daily average output up 1.6% month on month in July as Gulf plants resumed, Slovakia’s Slovalco restarts in Q4 2026, and Indonesia’s operating capacity is projected to reach about 2.51 million tonnes in 2026 from roughly 870,000 tonnes in 2025. Goldman Sachs raised its Q4 2026 forecast only to $2,500 per tonne from $2,350, seeing the surplus widening towards 1.5–2.0 million tonnes by 2027.

What the model misses

The framework treats warrant scarcity as a price signal, which is true only until it isn’t. LME nickel in 2022 showed that extreme concentration of deliverable stock produces a disorderly squeeze followed by exchange intervention, not a clean trend — an outcome that would deliver the price target and destroy the tradeability of the position simultaneously.

The second limit is restart optionality. Gulf capacity is not gone, it is offline. Emirates Global Aluminium and Aluminium Bahrain both disclosed damage from the March 2026 strikes, and SMM recorded only 89 of 1,262 pots operating at EGA’s Al Taweelah as of July 2. That restart curve is the largest swing factor in the balance, and it is a management decision as much as an engineering one. Against that, South32 placed its 240,000-tonne Mozal smelter on care and maintenance in March 2026, and Alcoa cut 2026 alumina guidance by 200,000–300,000 tonnes on Pinjarra instability. This is the same asymmetry that shaped our iron ore call for Q4 2026 and the platinum deficit-return case: a deficit that depends on an outage is a deficit with a repair schedule.

“Aluminum did not have the explosive price move that copper and tin had in 2025, and we think that will be the case this year.”

— Edward Meir, metals analyst and contributor, Aluminum Market Update (CRU Group, January 15, 2026)

Meir’s note predates the March strikes, which is precisely why it belongs here: it is the clean statement of the pre-shock consensus that aluminium is the base metal which does not spike, because idle capacity always answers the call. Anyone underwriting $3,600 is betting 2026 stopped being that year. His 400,000-tonne deficit estimate sits well below the 930,000 tonnes the bullish end of the sell side now carries.

What would invalidate this call

The base case to $3,600 breaks if ANY ONE of these four signals fires:

• LME total stocks rebuild above 400,000 tonnes. That would restore the inventory position to roughly where it stood on March 27, 2026, and remove the scarcity premium the entire thesis is priced on. The drawdown has been monotonic since late January; a sustained rebuild is a regime change, not noise.
• The Russian share of available LME stocks falls below 70%. A fall of that size implies non-Russian units are being delivered back onto warrant, which is the mechanical opposite of the deliverable-metal squeeze. It would signal that the US premium has stopped outbidding the exchange for non-sanctioned tonnes.
• EGA’s Al Taweelah returns above 900 of 1,262 pots before November 30, 2026. Gulf capacity is roughly 9% of global primary supply. A faster-than-guided restart converts the largest bull leg into a bear leg inside one quarter.
• LME cash posts two consecutive weekly closes below $3,085 per tonne. That level is the July 2026 low. Losing it with inventories at a 1998 low would say the market has decided the stock number is not the binding constraint, and the thesis is simply wrong about what is being priced.

What to watch next

Daily LME warehouse stock and cancelled-warrant reports are the highest-frequency read on the thesis; the cash-to-three-month spread flipping decisively into backwardation would confirm it, a return to contango would not. SMM’s monthly China production and casting-ingot split, published in the last week of each month, tests the liquid-aluminium leg. EGA’s restart disclosures and Alba’s potline updates are the swing variables, with quarterly guidance from Alcoa and South32 framing the ex-China balance. Any change to China’s 45-million-tonne cap — particularly a renewables-powered exemption, still under discussion — is the one policy headline capable of resetting the structure. Watch the US Midwest premium too: a decline below $1,800 per tonne would signal the tariff wall has stopped diverting metal.

Frequently asked questions

What is the current LME aluminium price?

LME cash aluminium settled at $3,327.50 per tonne on August 10, 2026, with the three-month contract at $3,320.50, according to LME official data republished by Westmetall. The metal traded as high as $3,336.50 intraday, its strongest since June 23. That represents a gain of roughly 10.4% from the January 2, 2026 level of $3,015 per tonne.

Why are LME aluminium stocks so low?

Exchange inventory fell to 254,900 tonnes by August 10, 2026 from roughly 445,000 tonnes in late January. Three forces drained it: Gulf outages after the March 2026 strikes cut output outside China by 6.7% year on year in July; the 50% US Section 232 tariff and a record $2,182 Midwest premium pulled non-Russian units to the United States; and Chinese smelters diverted 78.3% of output to liquid metal rather than warrantable ingot.

Does low inventory automatically mean higher aluminium prices?

No. Between early June and mid-July 2026 LME cash fell roughly 16% while stocks kept draining, because the market was unwinding a Strait of Hormuz risk premium faster than it was pricing scarcity. Inventory is a necessary condition for a squeeze, not a sufficient one. The bull case needs restarts to stay slow enough for the drawdown to become the dominant signal.

What does the falling alumina price mean for aluminium?

It cuts both ways. Alumina near $365 per tonne in mid-July 2026, against a Q4 2024 peak above $700, removes the marginal-cost support underpinning many bullish commodity theses. But it also widens smelter margins sharply, giving producers every incentive to restart idled capacity — the principal bear risk to a $3,600 target rather than an argument for it.

What do banks forecast for aluminium in Q4 2026?

Dispersion is unusually wide. Citi has carried a 0–3 month target of $3,600 per tonne with a $4,000 bull case, while Goldman Sachs raised its Q4 2026 forecast only to $2,500 from $2,350 and expects a widening surplus into 2027. ING has pointed to roughly $3,400 for Q4 2026, and Macquarie has forecast a 930,000-tonne deficit for the year.

How does aluminium compare with other industrial metals?

Aluminium is up about 29% year on year but lagged copper and tin through 2025 and remains below its March 2022 record. The distinguishing feature is inventory composition, not absolute price: no other LME contract has 93% of available stock concentrated in a single sanctioned origin. Our palladium analysis and USD/CNH call cover adjacent parts of the same complex.

This article is informational analysis only and is not financial, investment, or trading advice. Foreign-exchange, commodity, and equity markets are highly volatile and can lose substantial value rapidly. Leveraged products carry total-loss risk and may exceed the initial margin posted. Past performance and historical correlations do not guarantee future results. Do your own research and consult a regulated financial adviser before making any investment decision.