By Andy Home
LONDON, Oct 7 (Reuters) - You don't need to look very far to understand why nickel has been the consistent underperformer of the London Metal Exchange (LME) base metals pack since the middle of the year.
The explanation comes on a daily basis in the form of the LME's morning stocks report.
Today's showed registered inventory rising by a net 558 tonnes to 240,408 tonnes, an all-time record high - the latest in a long series of them - as surplus units spill into exchange warehouses.
Nickel is a market in chronic oversupply resulting from systemic over-production.
Supply needs to be cut if the market is to rebalance. But this market's supply profile is complex, with at least three moving parts.
The first is China's nickel pig iron (NPI) sector. There are still few signs that the expansion momentum in NPI is slowing. Rather, price pressures are forcing producers to switch to lower-cost technology, in effect reducing the collective cost curve.
That's put pressure on existing higher-cost producers in the rest of the world.
There has been some supply-side response but up to now it has been patchy. A few Australian mines have been shuttered. Glencore Xstrata has mothballed its ferronickel operations in the Dominican Republic, while Votorantim has done the same at its Fortaleza plant in Brazil.
Then there is the third wild card, the new projects currently ramping up with the sort of bad timing that commodity markets used to be famous for.
This third supply element is itself highly unpredictable, largely because four of these projects are working with high-pressure-acid-leach (HPAL) technology, which has a highly problematic track record.
And the operational challenges were again evident in the latest batch of quarterly production reports, with two downgrades to production guidance.
In the case of the Ambatovy project in Madagascar, this was the second consecutive downgrade.
Ambatovy, with nameplate capacity of 60,000 tonnes per year finished metal, was originally expected to produce 35,000 tonnes in its first full year of operation.
That figure was cut to 31,000 tonnes at the half-year stage and has just been cut again to 26,000 tonnes.
Sherritt International, majority owner and operator, said that ore throughput in the third quarter declined to 39 percent of nameplate capacity from 41 percent in the second quarter.
The core problem was "planned and unplanned maintenance activities" with the refining plant, "including acid injection system failures on Autoclaves 2 and 3 that caused external damage to the shell of the autoclaves".
Production guidance at the Ramu project in Papua New Guinea has also just been cut.
Highlands Pacific, a minority shareholder in Ramu, said in its Q3 report that the Chinese operator "has advised that they will not reach the targeted production of 15,500 tonnes of nickel (50 percent of nameplate capacity) for the 2013 year".
No new figure was provided but production through September was just 7,665 tonnes of nickel in hydroxide.
The Goro project in New Caledonia, or VNC as it has now been named by owner-operator Vale, is now in its third year of operation and has come to define the challenges involved in getting consistent performance out of HPAL.
Vale has just announced that September production of 2,200 tonnes of nickel in hydroxide and oxide was the "best month ever".
But quarterly production of 5,653 tonnes was only marginally better than Q2's output of 3,400 tonnes, while finished products output, once the intermediate products have been further refined by the company's plants in Taiwan and South Korea, fell to 4,700 tonnes from 6,600 tonnes.
Only First Quantum appears to have cracked the technology. Its Ravensthorpe operation in Australia produced a record 7,560 tonnes in mixed hydroxide in Q3 with 2013 guidance of 35,000-37,000 tonnes held unchanged.
Now, these continued operating problems and production downgrades might be viewed as a positive, given the evident oversupply in the nickel market.
But none of these HPAL operators is going to give up. Indeed the Q3 reports amounted to a collective statement of intent to continue working towards full capacity.
"The ramp-up of VNC showed significant advances, which make us confident of achieving performance targets for next year, when we expect to close the gap in cash flow generation," said Vale.
Sherritt said it is aiming for commercial production at Ambatovy, defined as consistent throughput of ore at 70 percent of design capacity over a 30-day period, in the fourth quarter of this year.
Ramu's operator, meanwhile, remains confident that the project will reach "full production capacity by the end of 2014".
Between them, these three projects are intended to deliver 150,000 tonnes of contained nickel per year. Cumulative production in the first nine months of this year was just 40,400 tonnes.
That means there is a lot more metal to come, if the operators' confidence is to be believed.
There is also a lot of more metal to come from two other projects that are in the process of ramping up.
Glencore Xstrata has started commissioning its 60,000-tonne per year Koniambo ferronickel project in New Caledonia, while Vale's 52,000-tonne per year Onca Puma ferronickel operations are expected to start producing again this quarter after a year's inactivity following a furnace blow-out.
The key point with all these projects is that operators need to keep pushing for full-capacity production to start recouping the massive capital costs.
There may well be more technical set-backs at the HPAL projects, given the evident challenges in achieving consistent operational performance, but the simple fact is that there are some 260,000 tonnes of capacity ramping up with incremental flow-through to a market already burdened with large and still-rising stocks.
The only real question is one of timing.
And the only real solution for this market, if it is to rebalance supply and demand, is the implementation of more and deeper voluntary curtailments.
Until that happens, today's fresh record high in LME stocks won't be the last.