There is certainly a precedent. When, during the Sixties Harold Wilson was fighting to protect a weak pound, Britain sold off a large portion of its gold reserves at the then official price of $33 an ounce - only to see gold surge to over $600 in the next 15 years.
This is, however, a weak precedent, because the gold was sold during the early stages of a period of great inflation, whereas now we seem to be heading into a period of prolonged deflation. Still, it was enough of a hunch to be worth exploring. Put at its highest, could this signal an end to the long bear market in gold? Put at its lowest, even if the bear market continues, could it signal a rally in the price? The short answer is that while the former does not seem very likely, the latter is quite probable at some time in the coming months.
With a commodity like gold the best guide comes from the very long term. The price over the last three-quarters of a century, both in actual dollars and in 1998 dollars, is shown in the graphic, taken from the latest issue of The International Bank Credit Analyst, which has tried to fit what is happening to gold now into its historical context.
Measured in 1998 dollars, gold was around $200 an ounce in the Twenties during the early stages of the post-war recovery, rising to around $400 in 1933-34 when reflation was at its height. It then gradually fell back to around $200 until the explosion of money supply, and subsequent inflation, in 1965-1980.
At the end of that period it reached over $1,200 an ounce; since when it has gradually subsided to below $300. But it is not yet back to $200, the level that seems to have been the floor this century. If disinflation continues, then you might expect it to continue down towards that point.
However even in the Eighties and early Nineties, during the switch from a world of inflation to one of price stability,central banks still felt they had to lean against inflation and there were periods when gold rose in price. These mini-rallies coincided with bouts of mini-reflation, in 1982-83, 1986-87 and 1993-95.
What happens next? Well, the world seems to be close to achieving price stability, certainly at a wholesale price level, if not yet at a retail one.
In most developed countries, wholesale prices have been negative over the past year. In some (Japan, Germany) wholesale prices have been sharply negative. True, that has been helped by very low commodity prices in general and true, the weighted average of world prices is still creeping upwards, but we must be quite close to an end of inflation. Put it this way. There has been a 20-year period when the long-term trend of inflation has been clearly downwards. There will not be another 20-year period like that, or even a 10-year period, for if there were, the global price level would be falling very fast indeed. A slow, long-term downward trend in prices is perfectly possible, but a sharp one cannot continue for long.
So there is a good chance that the world's central banks will ease up during the next couple of years. Sure, they may tighten policy this year: the Fed has already indicated that it has a bias towards tighter money; and while the European Central Bank may well make one cut in rates later in the summer, it too may become concerned about inflation in the second half and be tightening by the autumn.
But take a three-year view and it would be surprising were the central banks not switching back into a reflationary mode. The business cycle still exists. There will be a slow-down at some stage in the future. My guess is that there will be a post-millennial flop, but even if I am wrong, there will be a slowdown by 2003, and that slowdown will be met by efforts at reflation. When they do that, past experience suggests that gold will rally.
Is past experience still valid, given that central banks (and not just the UK's) are likely to be net sellers of gold for several years to come? The IBCA team reckons that it is, noting that during 1993, when central banks sold 20 million ounces of the stuff, the gold price still rose by $50. Of course the gold price, since it is expressed in dollars, to some extent is reflecting the strength of the dollar. While the dollar remains strong, which it looks as though it will for another six months at least, nothing very exciting is going to happen to the gold price. Dollar weakness, on the other hand, would naturally show in a gold rally, even if the price of gold expressing in other currencies did not move much.
There are also some wild cards. Were, for example, there to be unrest in South Africa, this would lead to a gold rally. Were there a sudden rise in demand for gold from the Indian sub-continent, ditto.
We tend to think of gold in terms of developed market demand, whereas in the years ahead what we still call the developing countries are going to have a larger and larger share of global output. Indeed, at some stage in the next decade it looks as if the total output of the developing world will pass that of the developed for the first time since the industrial revolution.
So the argument against gold, regarded as the killer punch in the West - why hold an asset which delivers no interest when you can hold one that delivers a real yield? - may become less relevant. In many developing countries currencies will remain distrusted.
None of this undermines the fact that in a world of deflation, gold will remain a lack-lustre investment: it loses its glitter. But knowing that there is a floor at $200 an ounce would be a start. And knowing there are rallies even in long-term bear markets would be a comfort to gold- bugs.