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

By Clyde Russell

LAUNCESTON, Australia, Nov 9 (Reuters) - - It's becoming harder to work out who are the most rational players in the global iron ore and steel markets, where contradictions are multiplying amid the persistent supply glut.

Iron ore miners and steel producers are likely to claim they are acting rationally, but both are probably guilty of selective thinking and a touch of amnesia.

Take Rio Tinto for example. The world's second-biggest iron ore miner reiterated last week that it has no plans to cut output amid the current oversupply, which has caused spot Asian iron ore prices to slump about 75 percent since the record high in February 2011.

At face value, this seems logical and rational. When you have the lowest cost mines in the world, why should you cut output, as this would simply allow higher-cost producers to take some of your market share.

"If you think for one second you can just take some volume out and no one else will actually move to fill that volume, then you are fooling yourself," Rio's iron ore boss Andrew Harding told a function in Perth last week.

But Harding's rational behaviour doesn't extend to recognising the reality that Chinese steel demand and output are most likely past their peak.

Rio insists on sticking to its view that China will reach peak steel production of about 1 billion tonnes by 2030, even though the China Iron & Steel Association (CISA), and virtually every other analyst, thinks last year's 823 million tonnes was the high-water mark.

Rio will argue that for its forecast to be accurate, China's steel output only has to rise at a rate of 1 percent per annum, but this ignores the current reality that 2015 consumption will see a decline in the region of about 5 percent from last year.

Iron ore imports by China are also now in negative territory for the first 10 months of 2015, following the 12.3 percent drop in October from September.

The question for Rio is then, while it's rational to maximise output even in the face of massive oversupply, how rational is it to stick to forecasts that are contradicted not only by the data, but also by the rest of the industry?

It's possible that China will experience a re-acceleration of steel demand in coming years, but this would have to involve a massive infrastructure programme, which isn't currently the policy or direction of the authorities in Beijing.

STEEL INDUSTRY BLINKERS

If Rio can be charged with selective rationality, then so can the Chinese steel industry.

The most logical, and profitable, thing to do when facing a situation of oversupply in steel capacity is to close down the highest-cost furnaces.

While some capacity cuts have been made, Chinese steel mills have generally chosen to continue to produce at a loss rather than trim output.

Steel producers suffered losses of 28 billion yuan ($4.4 billion) in the first three quarters of 2015, according to CISA, which added that overall capacity hasn't declined since 2010, despite the government issuing 20 policy documents calling for inefficient mills to be closed.

There is still the possibility that tighter credit conditions and soft demand will force the closure of some blast furnaces, but past experience suggests this will be a slow process and not nearly enough capacity will leave in order to restore the supply-demand balance.

For a left field example of contradictory thinking look no further than Lourenco Goncalves, the outspoken chief executive of U.S. iron ore and coal miner Cliffs Natural Resources , which also has a small iron ore mine in Western Australia state.

While lambasting Rio for its "entrenched" irrationality over its forecast for Chinese steel output to reach 1 billion tonnes, Goncalves also manages to state that Australia is destroying itself by supplying a finite resource cheaply to a China that may become an enemy.

"I think that Australia will only believe China is destroying Australia when they build an artificial island on the Great Barrier Reef that they can see from the shore," Goncalves was quoted as telling analysts in a Oct. 30 report in the Sydney Morning Herald.

While it's safe to dismiss Goncalves' more florid remarks, his comments that iron ore supply isn't leaving the market as fast as expected by the major miners, Rio, BHP Billiton and Brazil's Vale, are quite rational and accurate.

"In their imaginary world, 60 million tons of capacity will go offline this year, then another 125 million tons of capacity will go out of commission next year," Goncalves told Bloomberg in an interview published on Nov. 4.

"That's not the case. Everyone is driving down costs, everyone is trying to continue to cope. You're not seeing any meaningful number of tons going offline," Goncalves said.

While Goncalves analysis is rational, as is his conclusion that iron ore prices have further to fall, his preferred solution that the majors cut output to tighten the market is irrational as no major producer is going to give up market share to a smaller, higher-cost rival.

The most rational recent commentary on the iron ore and steel sectors has come from BHP's vice president of iron ore marketing, Alan Chirgwin, who told the Sydney Morning Herald that iron ore will find a price level at the highest breakeven of a major producer in Australia or Brazil.

This means iron ore will languish below $50 a tonne for several years, as both miners and steel producers first try to cut costs, and only cut capacity as a last resort.

What this shows is that in a market where everybody believes they are behaving rationally, it's still possible to get an irrational outcome.

(Editing by Richard Pullin)