Goldman Expects Aluminum Spike After Iran Attacks Damages The Two Largest Gulf Producers

On Saturday, both Emirates Global Aluminum (EGA) and Aluminium Bahrain (ALBA) reported drone attacks damaging smelting facilities after hits on Iranian steel infrastructure last week. Neither company (at the time of writing) have confirmed whether supply will be impacted, and until we have clarity on the production impact we can’t determine what level of risk-premium should be priced into aluminum on tomorrow’s Asia open, but Goldman commodity desk view is if we were to lose 600k MT, price should rally by 5-7% and prompt vol by 5%.

While Goldman commodity trader Adam Gillard acknowledges that (likely) growth downgrades from Houthis escalation and potential Bab-el-Mandan closure will limit the day-one rally, which also means spreads do part of the work, there is no obvious solver outside of demand destruction to this level of (potential) Middle Eastern supply disruption, and therefore flat price has to perform, especially given cleaner spec positioning, in Goldman's view.

Here are some more comments / numbers from the Goldman desk: 

• EGA: The damage was to Al Taweelah smelter (1.3mn MT) which has 3 pots producing ~ 450k MT each. We do not know which pots (if any) were damaged by yesterday’s attack. 
• Alba: Have 6 potlines producing 1.6mn MT. On 15th March pots 1, 2 and 3 (300k MT) were curtailed. Pots 4 & 5 produce ~375k MT each whilst pot 6 is ~540k MT. We do not know which pots (if any) were damaged.   
• Previous disruptions: Since the war started Qatalum have reduced output by 40% (280k MT) whilst ALBA have curtailed pot lines 1-3 (300k MT). We also think Iranian output has been reduced by 30% (200k MT) from damage to energy infrastructure. 
• Demand impact: GIR estimate a 600k MT 2026 demand loss from lower global GDP since the war began which in part explains why aluminium closed only ~3% above pre-war levels on Friday, positioning being the other reason (see below). Note GIR estimate 1% global GDP downgrade reduces aluminium demand by 1.9% / 1.4mn MT. 
• Inventory: LME is 420k MT (270k MT on-warrant) with another 100k MT off-warrant, whilst China has 1.4mn MT. Note there is a 15% export tax on Chinese primary aluminium which means for Chinese inventory to fill ex-China supply loss the import ARB needs to close (relatively easy) then the export ARB open – for the export arb to open the LME price has to outperform the Shanghai price by 15% (ceteris paribus premiums and FX etc. Note >70% of Chinese aluminium imports are from Rusal who toll via their stake in Hebei Wenfeng (so get a VAT discount). We struggle to see Beijing allowing energy exports (via aluminium given >40% of production cost is energy), in an energy crisis, though. 
• Positioning: We think aluminium struggled to price supply related risk premium in part due to stretched spec positioning; long aluminium was a very consensus trade (we had it also), almost everyone we know had ~7/10 length which explains the 8% sell-off as the market re-priced global growth. Anecdotally we think spec length has halved over the last two weeks because growth downgrades hit the complex more broadly - copper, gold (which is apparently a risk asset again) etc, but we’re struggling to explain why LME COT remains elevated. 
• Vol: We do not think the market is short top-side vol, nor do we think there is a significant local gamma short, but if price jumps by 10% think May25 rallies similar. Think cal-27 in the low 20s is too cheap (front-to-back on the lows); think $2mn vega is enough for the back to re-rate (but struggle to see where the supply comes from). 

Here, another Goldman commodity specialist, James McGeoch chimes in and says that he would use the recent Goldman research report "to understand the moving parts than for the price deck itself." That report, which is clealry dated now, saw the following Middle East alu production: 

According to McGeoch, client feedback has been buyers of dips.  He also notes that with regard to NHY, the stock has outperformed most global aluminium peers through March.

His parting warnings, "Remember - Demand shocks are linear, Supply shocks are often exponential."