Just WhenYou Think It's Over..

I've read some things written by gold bears of late, calling for gold to decline in March due to “seasonal tendencies."

As always with statistics, if one doesn't know why a particular statistical pattern occurs, one can make incorrect assumptions about future behavior.

In gold's case, it's typical seasonality tends to see gold rally in the fall, and then peak in the early spring. But the reason that we typically see this pattern, is due to the seasonality of gold demand coming from the largest gold jewelry consumer on the planet, whose various festivals revolve around gold: India and Saudi Arabia.

However, jewelry demand is not what is currently driving the gold price. Indian gold imports in February were virtually 0. Meanwhile, gold rallied $100 during February to $1000 and an 11-month high. As always occurs during a big bull market in gold, investment demand is the primary driver of the gold market, and it displaces jewelry demand. And investment demand is obviously not seasonal.

Thus, those expecting gold to follow the usual seasonal patterns may be in for a surprise. As I discussed in Countdown to the Gold Rally, the $100 decline from gold's peak of 10 days ago, was driven by an unusually high amount of short selling. I suspect this was based on the premise that the decline in Indian demand in February (and the "seasonal pattern") were predicting gold demand would wane, and lead to a decline.

If I'm correct, however, in my belief that investment demand is now the primary driver of the gold price (and not jewelry demand), then the gold market is set up for a vicious short squeeze. That squeeze could be augmented even further if the dollar declines sharply in the next few weeks, which gold -- as well as silver, platinum, and many other commodities -- appears to have been anticipating all along, with its rally since mid-December 2008 in the face of a rising dollar.

Meanwhile, the gold shorts continued to add to levered positions in the 2-times short gold ETFs this week - which short the COMEX futures to keep up their levered exposure. The shares outstanding of both ETFs below, which have exploded over the past few weeks as more and more bears have piled into these 2-times short ETFs, and pushed up COMEX gold open interest over the past 10 days - even as the price has declined.

Nevertheless, the GLD gold ETF (whose 150 tonnes of gold demand in February played a big role in pushing the gold price higher), hasn't seen its shares outstanding, fall by a single share. This means it hasn't sold a single ounce of gold since that $1000 peak in late February, either. Thus, anyone believing that the recent February rally in gold and the demand for the ETF was "retail" -- or merely, “fast money”-- has to seriously question that belief. The action in GLD thus far is, in fact, the classic definition of “strong hands” that don't react to price weakness by selling.

With gold having finally turned up yesterday -- and having added to those gains again today following the prior 8 straight days of heavy selling by shorts -- my bet is that the gold bears may be in for a big surprise, ishyak allah... But as always, we shall see.