By Andy Home /London


Another quarter and another record premium level for Japanese aluminium buyers. The premium over the London Metal Exchange cash price for third-quarter shipments to Asia’s largest importer seems to be settling around the $400 per tonne level, up from $365-370 in the current quarter. A year ago the premium was just $250 and until the third quarter of 2012, it had never been higher than $130.
Japanese buyers, however, have little choice but to take the premium pain.
Although still viewed as a benchmark for the whole of Asia, the Japanese quarterly premium no longer has anything to do with regional supply-demand nuances. It is now just one manifestation of a global premium structure.
If Japanese buyers baulk, producers can simply point to premiums in North America and Europe with the implicit threat of diverting shipments to those markets.
That, certainly, is the stark warning from Oleg Mukhamedshin, Deputy Chief Executive of Russian aluminium giant Rusal, speaking to reporters earlier this week.
“We think the premium can easily reach a new record high well above $500,” he said, adding for extra emphasis: “In the third quarter we can see new records, even $600 would not be out of the question.”
What once would have seemed a flight of producer fantasy doesn’t seem quite so far-fetched these days. Those who have tried to stand in the way of rising physical premiums have been steam-rollered - witness the short-covering frenzy that sent US Midwest premiums into supernova at the start of this year.
The US Midwest premium, as assessed by Platts, a leading global energy, metals and petrochemicals information provider, surged to an all-time high of 20.75 cents per lb ($457.45 per tonne) in early January. The pull-back made it only as far as 18.25 cents in early April and now the premium’s rising again, hitting 19 cents at the end of May.  The same is happening in Europe and the unstoppable premium machine has just struck Japan.
So just how high might premiums go? To answer that question would mean disentangling the various interconnected drivers. That is itself a major problem since there is no clear-cut consensus as to which of them holds the key to unlocking the premium riddle, a dilemma that was laid bare in the aluminium industry’s fractured response to the LME’s consultation on its warehousing policy.
Is it the load-out queues at LME locations such as Detroit and Vlissingen? Is it the in-vogue stocks-financing trade that has tied up so much surplus metal and kept it away from manufacturing users? Or is it the multiplying smelter closures that are pushing the market into deficit?
All are somehow in the mix and right now there’s still no easy way of saying which is the most important. Take those infamous load-out queues, for example. The LME’s proposed solution to its queue problem has already altered warehouse company behaviour even though its load-in-load-out formula is stuck in legal limbo thanks to the courtroom fisticuffs between the exchange and Rusal.
The revolving door strategy employed by Metro in Detroit, whereby rental from the load-out queue financed incentives to attract more metal into its sheds, has been stopped. Not one tonne of aluminium has been warranted in Motown in the last two months.
The load-out queue, however, has still been growing because metal is still being cancelled and joining the queue, 180,000 tonnes of it over the course of April and May.  As of the end of April, the aluminium queue at Detroit stood at 683 days. The cost of getting metal out, a combination of the daily rental and the load-out charge, was $388 per tonne, within a whisker of Platts’ then Midwest premium assessment of $406. Until the queue actually starts shortening, the apparent linkage between the two remains intact. On current trends, the day is fast approaching since there are now only 158,525 tonnes of non-cancelled aluminium in the city.
But then there is Vlissingen, the Dutch port dominated by Pacorini, the warehousing arm of Glencore. Although inflows have slowed, they have by no means stopped and it looks like movement is being carefully calibrated to minimise queue decay in the event the LME’s new policy makes it through the courts.
The aluminium queue here was 748 days at the end of April and the notional “value” was $406 per tonne. For those, and they are many, who argue that physical premiums are all about queues, there’s nothing here to prove them wrong yet, although things will get a lot more interesting when the Detroit queue starts contracting.
But it should be clear by now that queues are not the only driver of premiums. After all, if less metal is flowing into the LME warehouse system and more leaving, why hasn’t there been an impact on physical market availability and therefore premiums?
The answer comes in two parts.
Firstly, hardly any of that metal leaving Detroit and Vlissingen every day is going anywhere near an actual manufacturer. Rather, it is still mainly, and quite possibly totally, going to cheaper off-market storage to earn a tidy profit for stocks financiers.
The trade remains in robust good health. True, there has been a sharp contraction in the front part of the LME forward curve over the last couple of days, the benchmark cash-to-three-month period closing Wednesday valued at $22.25 contango, compared with over $40 a week ago. But we’ve seen these spread spasms come and go over recent years with little impact on the bigger financing picture.
And although the prospect of higher interest rates is looming larger in the US and the UK, the European Central Bank has just moved in the opposite direction with a further loosening of monetary policy. As long as money remains cheap and interest rates low, one of the pillars of stocks-financing profitability remains firmly in place.
Then, of course, there is the fact that there is less metal around to plug the supply gap left by the financiers. As more and more smelter capacity is mothballed, this fundamental driver of higher premiums is assuming ever greater significance.
North American production was running at an annualised rate of 4.6mn tonnes in April, according to the latest figures from the International Aluminium Institute. That’s the lowest level since August 2010.
Smelter capacity has been permanently lost in both western and eastern Europe over the last couple of years with many plants in the former still struggling to survive.

The opinions expressed are those of the author, a columnist for Reuters.