Alcoa is something of a corporate bellwether for the global aluminum industry.
The US company's operations span the full gamut of the light metal's value-added chain from mining the bauxite needed to produce the stuff all the way through to specialist engineered products tailored to meet the requirements of aerospace and automotive customers.
Its Q2 results, therefore, provide a telling snapshot of the strains on the entire aluminum sector resulting from the current price weakness.
On the London Metal Exchange (LME) benchmark three-month aluminum hit a one-year low of $1,832 per ton in June. As with the other base metals traded on the LME an early-July rally has quickly run out of steam and aluminum is once again foundering around the $ 1,900 level.
So no particular surprise as to the prime cause of the slide in Alcoa's headline earnings from net income of $ 94 million in the first quarter of the year to a net loss of $ 2 million in the second.
While downstream divisions performed well, and with rolled products even generating record operating income in the first half of the year, Alcoa's primary metals business took a $ 58-million sequential hit from the lower aluminum pricing environment.
There was some clawback from currency effects, $ 19 million, and from higher physical market premiums, $ 15 million.
But not enough to stop the division sliding into the red in Q2 to the tune of a net $ 3 million.
All of which palpably frustrates Klaus Kleinfeld, Alcoa president and chief executive officer.
After all, Alcoa is still projecting global demand growth of 7 percent this year, split between 11-percent growth in China and 3 percent in the rest of the world.
If even close to the mark, this would be a remarkable outcome given the combination of engineered slowdown in China and far-from-engineered slowdown just about everywhere else.
But then aluminum is leveraged to key growth sectors such as automotive, where it is gradually winning a long-term materials battle with steel, and aerospace, characterized by an order backlog stretching to years and an ageing commercial airline fleet.
Stocks of aluminum, according to Alcoa, may still be high but at 76 days' global usage they are down on the 102-day peak seen in 2009 at the height of the Great Contraction.
The price of aluminum, argued Kleinfeld on the Q2 analysts call, has decoupled from this fundamental demand strength, becoming subsumed within the general macro risk-on-risk-off trade that has dominated all asset classes ever since.
That is at least partly true. Many base metals have become more closely correlated with other asset classes over the last four years.
But there is the thorny question of whether the world is still producing too much aluminum, which could of course be one fundamental reason why the price is where it is.
Alcoa thinks not, forecasting a global 515,000-ton production deficit this year.
It's a nigh unprovable proposition given the opacity of the aluminum sector, not least in China, both the world's largest consumer and producer.
In short, you're going to have take Alcoa's word on it. And Alcoa, in Kleinfeld's words, is going to have to "let the market do what the market is going to do."
Since the market is stubbornly declining to reflect Kleinfeld's optimistic analysis of aluminum's "sound fundamentals," Alcoa's ongoing response is something of a survival handbook for every other margin-compressed producer.
Alcoa is "targeting every single lever that we have" to improve group productivity from stream-lining purchasing to more efficient scrap usage to maximizing products capability to reducing days' working capital.
But at the core of the strategy is the company's five-year plan to move 10 percentage points down the global aluminum cost curve from 51st percentile in 2010 to 41st in 2015.
And here smelter portfolio management is key.
If smelters have been idled for several years with diminishing chances of ever reopening, dismantle them.
Alcoa has permanently closed 291,000 tons of capacity, comprising all of the Tennessee smelter and one-third of its Rockdale smelter in Texas.
If smelters are not competitive at lower prices, idle them. Alcoa is in the process of mothballing some 14 percent of its global metal capacity, mainly in Spain and Italy.
Such action can, of course, be presented to the wider world as taking global leadership in response to difficult market conditions.
And if you curtail primary smelter capacity, you may as well curtail alumina capacity, as Alcoa has done with 390,000 tons of its intermediate product capability.
If government levers can be used to preserve struggling smelters, use them.
Alcoa has been reviewing the future of both its Point Henry smelter in Australia and its Brazilian smelter operations.
Point Henry has just been given a stay of execution until at least mid-2014. An assistance package worth more than A$40 million from the Australian government may have helped.
In Brazil Alcoa "has raised the issue of competitiveness, or lack of it, on energy prices," the single more important input variable in any smelter's profitability profile.
Kleinfeld told analysts that "I believe there are going to be actions taken sooner rather than later" by the Brazilian government. And finally, if you can't beat new lower-cost entrants to the global aluminum sector, join them.
Alcoa is building an ultra-low cost integrated alumina-aluminum-products facility in Saudi Arabia.
The 740,000-ton per year smelter component of the Maaden project is on schedule to start production next year.
This is the law of the jungle. Remember the old joke. You don't need to outrun the lion, just the guys running with you.
The only problem is the cost curve itself.
The most obvious point is that it is never static, largely because everyone is simultaneously trying to move down it, redefining it as they do.
This is what is happening in China with smelter capacity migrating from east, where power availability is restricted and prices accordingly high, to west, where stranded coal deposits afford much lower power prices.
And it's happening elsewhere. Alcoa is not the only producer migrating capacity to the Middle East to tap into the region's abundance of energy. Rio Tinto and Hydro have already done so with new smelters in Oman and Qatar respectively.
A more fundamental issue with cost-curve economics in the aluminum market, though, is the gap between theory and reality.
If governments are prepared to subsidize power prices, as is the case with an estimated 25 percent of Chinese aluminum capacity right now, then evidently the cost curve is not what it should be on paper.
This isn't just about China, witness the Australian government's helping hand to Alcoa's Point Henry smelter and the Brazilian government's consideration of the "competitiveness, or lack of it" in local energy prices.
The resulting problem is, to quote researchers at Barclays Capital, that such "interventions disrupt efficient market mechanisms resulting in distortions and dislocations with bearish implications for fundamentals and prices."
Government assistance negates the cost curve at precisely the point it should be effective in matching supply to demand.
As a consequence, as BarCap points out, the market will remain prey to over-production resulting from the build-out of excess capacity.
You may well believe Alcoa's contention that enough capacity has been cut to generate a market deficit this year.
If you don't, however, the implication is that prices must go lower to compensate for non-market intervention.
And what will the cost curve look like by 2015? Will Alcoa have achieved its goal of being in the 41st percentile? Or will it need a new five-year plan to replace the current one?
This chase down the aluminum cost curve is one without obvious end. Alcoa must keep running. So too must every other producer.
- Andy Home is a Reuters columnist. The opinions expressed are his own.