COLUMN-Is the curtain coming down on U.S. aluminium smelting? -Andy Home

(Repeats NOV 24 column, no change to text)

By Andy Home

LONDON, Nov 24 (Reuters) - At the end of the last century the United States was home to 22 aluminium smelters, all but one of them operating.

By the end of this year there will be just eight, of which only four will be producing metal, two of them at reduced rates.

The latest round of closures, led by Alcoa (NYSE: AA - news) , is happening just as the metal's usage in the United States is set to experience another quantum leap forward.

Aluminium has made steady inroads against steel in the automotive sector, a process that is going to markedly accelerate with the roll-out of the mass-market F-150 pick-up truck.

Number one on the list of "the tough 10 reasons every other truck is history", according to Ford's advertising campaign, is the fact the F-150 has a "high-strength, military grade aluminum alloy body."

It (Other OTC: ITGL - news) is a bitter irony for U.S (Other OTC: UBGXF - news) . smelters that the low price environment that is forcing them out of business only adds to the attraction of incorporating aluminium into applications such as the F-150.

But the simple reality is that on a global basis there is too much of the stuff around.

That's why Alcoa's announcement this month that it is shuttering three out of four of its remaining smelters in the United States has generated so little lift to the aluminium price.

However, it does push the U.S. market into deeper supply-demand deficit, a development likely both to underpin regional premiums and heat up the exchange battle for trading those premiums.

AND THEN THERE WERE...

Alcoa's announcement that it is mothballing another three of its U.S. smelters leaves it with just one operating plant in the country, Warrick in Indiana, with a capacity of 270,000 tonnes per year.

The shuttering of the Wenatchee and Intalco smelters in the state of Washington and the Massena West smelter in New York will remove around 500,000 tonnes of production.

It is part and parcel of Alcoa's drive down the cost-curve with a target of moving from the 43rd percentile this year to the 38th percentile next year.

This has been work in progress for Alcoa for some time but extra urgency comes from the company's plans to split into two entities next year.

The "old" Alcoa, comprising the upstream business of mining, refining, smelting and casting metal is being moulded into "a commodity business that is positioned to succeed throughout the cycle", to quote Klaus Kleinfeld, chairman and chief executive officer.

With (Other OTC: WWTH - news) the exception of Warrick, which benefits from its own captive coal power, its U.S. smelters apparently are being relegated to "swing capacity" status.

A similar process is underway at Century Aluminum (NasdaqGS: CENX - news) , a U.S. producer majority owned by Glencore (Xetra: A1JAGV - news) .

It owns three plants, not counting the Ravenswood smelter which was mothballed in 2009 and which is now to be permanently decommissioned.

All three are in trouble.

Operations at the Hawesville smelter in Kentucky have been reduced to two lines, or around 40 percent of the plant's 250,000-tonne per year capacity, to produce high-purity metal for local customers.

One of three potlines at the Sebree smelter, also in Kentucky, has just been mothballed, taking out around 70,000 tonnes of capacity.

Mt Holly in South Carolina, meanwhile, is scheduled for full mothballing by the end of this year unless a new power supply deal can be negotiated.

The only other U.S. smelter operating is the New Madrid plant in Missouri, part of Noranda Aluminum (NYSE: NOR - news) .

Noranda too, though, is feeling the heat, judging by its third-quarter results, which saw the company slide into a $26.4 million loss and write off any remaining goodwill in its smelting operations.

GLOBAL PROBLEM

U.S. smelter capacity has been bleeding away since 2001, when a cluster of west coast smelters shut up shop.

The process has been almost continuous since the global financial crisis of 2008-2009, which saw prices collapse and inventory mushroom as producers failed to react quickly enough to the implosion in manufacturing demand.

While China's massive infrastructure spend of 2009-2010 helped other metals such as copper, it failed to generate the same recovery in aluminium, because China itself was already a massive producer.

Indeed, the pace of new smelter capacity build-out in China's northwestern provinces such as Xinjiang has steadily increased to the point that the country now produces over 50 percent of the world's commodity-grade aluminium.

And it has been exporting ever larger quantities in the form of semi-manufactured product, some of which has challenged the definition of what exactly is a "product" rather than metal transformed just enough to duck China's 15-percent export duty and instead qualify for a VAT rebate.

"Fake semis" are a hot political subject in the United States right now with Century lobbying hard for action against what it sees as a flood as mislabelled China product, some of it from Chinese smelters enjoying local government subsidies in the form of reduced power tariffs.

But notwithstanding such concerns, there is the unpalatable fact for U.S. smelters that in an oversupplied global market-place they cannot compete with the new generation of smelters in China's northwest.

Sitting on their own captive coal fields, these plants benefit from low-cost power, state-of-the-art technology and efficiencies of scale. None of them are subsidised because they have no need for subsidies.

It's for that reason that the price has been steadily sinking to a point that at a current $1,450 per tonne, basis three-month metal on the London Metal Exchange, it is close to the troughs of 2009.

MADE IN CHINA?

The longer-term problem for the aluminium price is not current over-production but excess capacity, particularly in China.

While Alcoa and other U.S. producers have been willing to dismantle idled smelters after a few years of inactivity, there seems to be no similar process in China.

The fear is that any turnaround in price will be capped as such smelters fire up again and there's a lot more "swing capacity" in China than Alcoa's 600,000 tonnes or so.

But the decline and fall of U.S. smelting capacity should support one component of the price, the "local" Midwest premium, as the country becomes increasingly reliant on imports to meet its burgeoning demand for raw metal.

Part of the growing supply-demand gap will continue to be met by Canada, where smelters, including three owned by Alcoa, enjoy the power-price benefits of Quebec's massive hydro-electric system.

But more of it will have to come from further afield, from Russia and the Middle East, implying higher freight rates.

U.S. smelters' loss may turn out to be exchanges' gain with CME's Midwest premium contract already gaining traction and the LME this week launching its own set of premium contracts.

For U.S. manufacturers that may be an increasingly important hedging tool as the country's ability to supply its own needs diminishes.

And who knows?

Might there come a day when the Ford F-series pick-up, the "best-selling brand of trucks" in the United States (No. 10 on the list of "tough 10 reasons every other truck is history") is made from alloys cast from metal smelted in China?

(Editing by Jason Neely)