LONDON (Reuters) - A semblance of normality has returned to the aluminium market after three weeks of turmoil caused by the imposition of U.S. sanctions on Russian oligarch Oleg Deripaska and his Rusal empire.
The extension of the sanctions deadline until Oct. 23 and the potential for some sort of political deal have helped calm troubled waters.
The London Metal Exchange (LME) three-month aluminium price hit a seven-year high of $2,718 per tonne on April 19 but is now trading back down around the $2,250 level.
The same, however, cannot be said of the raw materials market.
Alumina supply was already tightening thanks to the partial, continuing outage at Hydro’s Alunorte plant in Brazil.
The prospect of losing Rusal material as well sent prices stratospheric, revealing both the fragile nature of the alumina supply chain and its dysfunctional pricing model.
This is an industry that has given up its natural price hedge but not yet embraced a new pricing mechanism.
The events of the last three weeks have shown just how dangerous that can be.
Even before the April 6 announcement of sanctions on Rusal, the alumina supply chain was struggling.
In February a Brazilian court ordered a 50 percent cut in output at the Alunorte refinery following allegations of a waste spill. That ruling has just been upheld by a local judge.
Alunorte is the world’s largest single converter of bauxite to alumina and the continued outage translates to an annualised supply hit of almost three million tonnes.
When the news of sanctions against Rusal broke on April 6, the initial reaction was registered in the aluminium price, both the LME basis price and U.S. physical premiums.
It was only after a few days that the market realised just how vulnerable to sanctions were Rusal’s overseas bauxite and alumina operations.
And how vulnerable Western world aluminium smelters are to sudden changes in the raw materials supply chain.
At one stage there was a very real prospect of Rio Tinto halting shipments of bauxite to Rusal’s Aughinish refinery in Ireland, which in turn could have caused several European aluminium smelters to close.
That nightmare scenario has been deferred, but not completely dispelled.
The industry has got used to existing on lean inventories, some refineries holding less than one week’s operating stock, and ignored the increasing complexity of alumina trading between big entities such as Rusal.
On paper Rusal is almost self-sufficient in alumina. In practice, it trades a significant part of that material. Sales to third parties totalled 2.3 million tonnes last year, compared with group production of 7.8 million tonnes.
Alumina flows are accordingly more intricate and less transparent as producers and traders switch and swap tonnages across regions and time-frames.
(Graphic on aluminium and alumina relative price performance: tmsnrt.rs/2remUIh)
The alumina price has shown no indication of tracking the aluminium price lower.
The CME spot alumina contract, which is linked to the S&P Global Platts alumina price index, is holding the recent highs, closing Tuesday valued at $643 per tonne.
It’s still playing catch-up with the physical market, where prices have surged to $700 and above in the last couple of weeks.
The combination of a high alumina price and low aluminium price crushes operating margins for those smelters which do not enjoy their own vertically-integrated feed.
And few of them have any price hedge.
The industry used to link its alumina pricing to the LME aluminium contract, an unsophisticated but largely effective way of cushioning input costs against the sort of metal price blow-out we’ve just seen.
But around the turn of the decade Alcoa led a producer shift to pricing alumina on the basis of spot market indices compiled by price assessors such as Platts, Metal Bulletin and CRU.
Alcoa is now pricing around 95 percent of its third party alumina sales basis either an index or the spot market, it said in its Q1 results.
Most of the other big players in the market have followed suit.
There are two problems here.
The first is the nature of spot market pricing.
Speaking at last week’s CRU aluminium conference in London, Anthony Everiss, head of the group’s alumina and bauxite research, conceded that the record highs seen in the CRU index were based on just three or four cargoes, each around 30,000-35,000 tonnes in size.
This is a highly marginal tonnage in the global alumina market and symptomatic of the problems facing all the companies attempting to assess the alumina price.
The spot market is still too small and too China-oriented to offer a truly global pricing mechanism.
And the CME contract, based on one of those indices, is itself still too small to offer any practicable hedging opportunity.
Launched in 2016, it has yet to build any critical trading mass. Cumulative volumes in the first quarter were just 87,500 tonnes, less than the handful of spot cargoes used to calculate the underlying indices.
Yet on such insignificant tonnages turns the global alumina price and with it the operating margin for a significant part of the Western world’s smelter system.
The LME is itself planning to launch an alumina contract either late this year or early next year, in theory doubling the number of hedging venues.
But, like the CME, it too will be using a vanilla futures contract template with pricing indexed back to one of the core alumina price assessors.
The underlying issues of how those prices are set in the physical market and low liquidity in the paper market remain.
What really needs to happen is for producers such as Alcoa and Rio Tinto to extend their enthusiastic embrace of non-LME pricing to committing the tonnages to make that pricing work in trading venues such as the CME and LME.
They have just been given a major wake-up call about the broken nature of the current alumina pricing model.
They will ignore it at their own peril and that of a good part of the aluminium smelter sector.
The opinions expressed here are those of the author, a columnist for Reuters.
Editing by David Evans