COLUMN-Tumbling aluminium premiums put LME queues back in spotlight: Andy Home

(Repeats May 13 item. The opinions expressed here are those of the author, a columnist for Reuters.)

By Andy Home

LONDON, May 13 (Reuters) - The London Metal Exchange (LME) has announced another change in its warehousing rules to force a faster reduction in the load-out queues that have caused so much controversy in the aluminium market.

Its new regime of linking load-in to load-out rates, LILO for short, is now in force, albeit after a near one-year delay caused by a failed legal challenge from Rusal (HKSE: 0486-OL.HK - news) .

LILO is intended to choke off new supplies to those warehouse operators that might be tempted to rebuild falling queues.

However, the LME's own modelling suggests the most problematic queue of all, that at the Dutch port of Vlissingen, could still take over four years to fall below the targeted threshold of 50 days.

So the exchange has tweaked the LILO formula to force a faster decay rate. Once the amendment kicks in from August, that waiting time will fall to just over two years, even allowing for maximum flexing of the model by Pacorini, the "owner" of the Vlissingen queue.

Two years is still a long wait to get metal out of the LME system but with aluminium physical premiums collapsing the world over, the furore over load-out queues has lost a lot of its heat.

After all, it was the perceived linkage between queues and premiums that caused so much wailing and gnashing of teeth among aluminium buyers.

As LME price and "all-in" aluminium price rapidly reconnect, buyers have increasingly less to worry about.

Or do they?

With premiums falling faster than queues, might the linkage between the two return?

****************************************************** Graphic on LME aluminium load-out queues: http://link.reuters.com/vaj74w ******************************************************

DETROIT WIND-DOWN

The latest LME queue data shows three warehouse operators with load-out times above the 50-day target.

Originally there were five but queues at Antwerp and Johor disappeared many months ago.

There remains a queue for zinc at warehouses operated by Pacorini in New Orleans. But it is a small one, just 56 calendar days at the end of April.

That leaves what the LME calls the two "embedded queues" at Detroit and Vlissingen, where waiting times for aluminium ended last month at 406 and 474 days respectively.

Detroit was where the original aluminium load-out queue first appeared with Metro (Other OTC: MTRAF - news) , the dominant LME logistics player in Motown, credited with masterminding the queue-creation model.

However, under both former owner Goldman Sachs (NYSE: GS-PB - news) and new owners the Reuben brothers, Metro seems to have left the queue business.

Metro warehouses in Detroit haven't received new metal of any kind since July last year.

That renders LILO redundant at this location and, unless inflows reappear, the queue can be expected to tick steadily lower over time.

Indeed, the amount of LME metal stored at Metro Detroit has halved over the last year and in February fell through the 900,000-tonne level, reducing the company's minimum daily load-out requirement from 3,000 to 2,500 tonnes per day.

ALL EYES ON VLISSINGEN

Pacorini Vlissingen, on the other hand, has shown every sign of wanting to preserve its queue potential.

The queue at Vlissingen, where Pacorini was storing all but 618 tonnes of the 1.85 million tonnes of LME-registered metal at the end of April, remains highly dynamic.

Unlike Detroit, Pacorini has been loading in significant quantities of aluminium, almost 263,000 tonnes since the start of this year.

Vlissingen has also seen high levels of cancellation and reverse cancellation activity, suggesting the queue here is still being actively "traded".

As of Wednesday, Vlissingen holds 1.82 million tonnes of registered aluminium with a near 50:50 split between material on warrant and material on cancelled warrant queuing to exit.

What that means is that there are still over 929,000 tonnes of warranted aluminium that could in theory cause the queue to lengthen again.

Metro Detroit, by contrast, held just 105,545 tonnes of LME warranted metal at the end of last month, a much smaller feed of potential fuel for queue creation, even if Metro were so minded.

LILO, in both its current form and its foreseen accelerated form, is really all about Vlissingen.

Indeed, the whole LME campaign against load-out queues is increasingly being distilled down to a stand-off with Pacorini and its owner, Swiss commodities giant Glencore (Xetra: A1JAGV - news) .

QUEUE ARBITRAGE

The controversy over LME queues has faded over the last year or so, partly because of the exchange's evident determination to resolve the issue but also because any linkage between queues and physical premiums had broken.

Physical premiums were closely correlated with queue length a couple of years ago, although Goldman Sachs, Metro's owner at the time, argued vociferously that correlation did not mean causation.

Most aluminium users disagreed.

The argument was never settled before premiums surged at the beginning of 2014 to levels that were far and above anything that might be explained by queue length alone.

That bubble, however, is now popping with premiums in near free-fall just about everywhere.

Indeed, they have fallen so far that, on paper at least, it would be profitable for a warehouse to use the revenue from its queue to bid for fresh aluminium supply.

The revenue value of the two queues at Detroit and Vlissingen is around $270-280 per tonne, based on daily rent and load-out charges. (Detroit may have a shorter queue than Vlissingen but Metro's rental of 54 cents per tonne per day is higher than Pacorini's 50 cents.)

The CME's spot contract for the Midwest U.S. aluminium premium, which is indexed to Platts' assessment of the market, is trading at 10.25 cents per pound, or around $226 per tonne.

The hard arbitrage between queue length and physical premium, in other words, is once again profitable.

Assuming, that is, either Metro or Pacorini wants to try another variation of the old queue-premium game.

Any attempt to do so would mean greater load-out requirements under the LME's LILO rule and invite high levels of scrutiny from the exchange.

One of the many new rules the LME has introduced to try to resolve its warehousing woes is greater investigation powers over incentives paid by warehouse operators to attract metal into their sheds, particularly those operators with queues.

But then again, where there's an arbitrage, markets tend to find a way.

The aluminium market has resembled a giant commodity pricing experiment ever since the global financial crisis, when all that aluminium first started surging into LME warehouses.

The arguments about the last phase of the experiment, specifically to what extent lengthening queues caused rising premiums, are still reverberating around the industry.

The next phase of the experiment, whether queues can place a floor beneath premiums, is only just starting. (Editing by Dale Hudson)