Gold prices at $6300?

No doubt, it'll happen some day. Even if the price of gold stays constant in real terms, if you believe the Federal Reserve can stick to a 2% annual inflation target, that $6300 ought to roll around by, um.. the end of the century, give or take a year or two.

OK, so actual U.S. inflation is likely to run higher, to judge by the past century--when it's averaged 3.3% per year. In which case gold ought to hit the level by 2065.

But that's not what Societe Generale's Dylan Grice is arguing. He figures gold prices could ramp up to $6300 or more within the decade and possibly over the coming few years.

Unlike many gold bugs, Grice accepts gold's limitations. Historically there have been long periods (measured in hundreds of years) when it's lost value in real terms. It has no commodity value. And it generates no yield. Indeed, there's a negative carry if you figure you have to insure the stuff against theft or pay the cost of storage.

Instead, he figures gold is the next asset due to benefit from a mania-driven bubble. And he gets his target price from where the metal would need to be valued to ensure the U.S.'s total monetary base (the Fed's narrowest measure of how many dollars there are circulating in the economy) is fully backed by gold.

For the purpose of comparison, during the past 40 years, the price of gold has only briefly been at or above where it covered the monetary base--in the late 1970s/early 1980s, when it spiked sharply at the end of the great inflation.

Is this a reasonable metric to use for valuing gold's potential? Not really, as Grice admits. But it's also no stupider than some of the stuff people were using during the Nasdaq bubble. The point is, when a mania hits, people will latch onto anything to justify their enthusiasm. And Grice thinks a mania is in the offing.

The reason, he figures, is that the massive expansion of the Fed's balance sheet, alongside every other central bank's, together with enormous government deficits across the world, will drive considerable inflation. Governments and central banks won't have the courage or political backing to dole out the harsh medicine that'll be needed to control this inflation for some considerable time.

Is he right?

Well, investors certainly seem to be worried, to judge by how risk assets have raced away during the past six months or so. There's a sense central bankers will welcome a sudden surge of inflation. Plenty of economists are arguing that a hefty dose of inflation (say in the upper single digits) is exactly what the doctor ordered for debt-overloaded Anglo-Saxon economies.

But to judge by Japan's example, generating that sort of inflation could be trickier than people think. The Japanese have continually struggled with deflation during the past 20 years and, if anything, it seems to be getting worse. Which is one reason why, despite the recent jump in gold prices, in yen terms they are still only half of where they were during the late 1980s.

Is gold a particularly good hedge against inflation, though? Not really. The correlation between 12-month changes in gold prices and 12-month U.S. consumer price inflation since the start of 1969 has been a modest 43%. Nor does gold seem to anticipate generalized inflation--the correlation between gold price inflation and consumer price inflation one and two years down the line is approximately nil.

Changes in the gold price seem to anticipate changes in bond yields--there's a 54% correlation between 12-month changes in gold prices with changes in 10-year U.S. government bond yields over the following year.

So maybe what's happening in the gold market now is a warning for what will happen to bond yields in the coming months. But the relationship doesn't tell us much about where gold might be heading.

Could gold move even higher? Sure. It could also fall out of bed. And without any yield with which to console themselves, investors would feel the full force of such a blow.