Trader’s Narrative

The US dollar has fallen to its long term support. And I do mean long term (see chart below).

Since the early 1990’s it has approached this level 5 other times. In all but one case, it has reacted by bouncing off support.

The exception was in 1992 when for a short time the index breached the 80 “line in the sand”. But it then bounced above and went on to rise much higher. A classic bear trap.

Right now we are still above 80. History is our guide but anything can happen in the markets. We could bounce higher, we could go lower or we could just sit here at these levels for a while.

What I’m going to be watching is sentiment. Keep a watchful eye for negative articles or even better headlines and cover stories about the dollar. Or about the debt held by China.

The best contrarian signal the longs could hope for would be a very negative cover story in a general interest magazine like Time or Newsweek. Maybe Business Week will pull through as it has so many times.

Since the dollar and gold are intricately linked, any bounce for the dollar here will be the final nail in the coffin for the gold sector. But then again, that massive consolidation in the Phili Gold index (HUI) could turn out to be a bull flag. I doubt it. But who knows.

Click to Enlarge Graph:

There is an inter-market relationship between bonds and gold which may not be evident at first glance. The reason why the two disparate markets are connected is that they both rely on one factor: inflation. Or the expectation of future inflation.

For this reason, it is very rare for these two markets to go in different directions in the long run. Instead, they tend to dance around each other.

When the market expects future inflation to be higher, gold mining stocks rise and bond prices fall (yields rise). So in effect, by watching one market, we can attempt to gain insight into the other.

Jay Kaeppel developed a simple system to do just that. He looked at the 12 month rate of change (ROC) in the GMI (Barron’s Gold Mining Index) and compared it to the 12 month performance of bond prices going forward. We can substitute the Amex Gold BUGS Index (HUI) for GMI.

When the ROC for the gold index is positive and high, we would expect bond prices to decline going forward (next 12 months). Conversely, when the rate of change for HUI is negative or low, we’d expect it to be a great time to buy bonds. And this is generally what happens according to historical data.

Be sure to check out Kaeppel’s original article on this: Maximizing Bond Fund Profits. You can find it in the Reports & Articles section of my free trading resource “goodies box”.

So what is this simple intermarket relationship telling us now?

Well, as far as I can tell, the bond market and the gold market have decoupled. I can’t seem to find any meaningful relationship between them now. Kaeppel’s article was written in 1994 so it only included data until then. By looking at data from the late 1990’s to today, it seems that the two markets have gone their separate ways. Atleast for now.

For example, in the summer of 2002 when the 12 month ROC for the HUI index was astronomically high, bond prices actually were higher a year later. And in the summer of 2005 (see graph below), when the 12 month ROC was negative (-40%), bond prices actually peaked and were lower a year later.

Click to Enlarge Graph

Strange. As I’ve mentioned before, although the bond market is pricing in future expectations of inflation, other markets, like gold, silver, and the CRB commodities index are breaking down and not confirming this. What to make of this?

On April 25th 2007, I wrote that gold bulls would be disappointed (again). Lets take a look to see what happened since then and what we can tell for the sector going forward.

I pointed to the Commitment of Traders report, which showed the “smart money” commercials being markedly net short. Since then that lopsided situation worked itself out.

On April 17th 2007, the commercials position was: long 79,882 and short 254,480, which is a net short position of 174,598 contracts. From the previous week (April 10th 2007) the commercials had reduced their longs (-7,564) and added considerably to their shorts (+13,830). Meanwhile, the small speculators were long 54,973 and short 15,733, for a net long position of 39,240 (on April 17th 2007).

Now compare that to the COT on May 29th 2007. The commercials had a long position 125,989 of and a short position of 244,684 for a net short position of 118,695 contracts. And the small speculators were long 57,428, and short 33,630 for a net long position of 23,798 contracts.

An important metric to watch for the COT is the open interest. Currently it is at 425,688 contracts, a 12 month high. Usually important changes in trend develop when the market is positioned lopsided (commercials vs. speculators) and when the open interest reaches significant levels. We saw this happen in late February 2007 where the commercials had been increasing their net short position while the small and large speculators were going more more long. With the open interest at around 415,000 contracts, things hit their climax and it started to unwind.

As well, the Rydex sentiment measure was flashing a caution sign. Considering also that the gold index fell to an area of previosu support, it wasn’t surprising to see it rally last week. It started with hammer on Wednsday and then two back to back long range up days:

Last week may have gotten the gold bugs rejoicing. But when you step back and look at the larger picture you see a very lackluster performance. Relative to the general stock market, the gold sector has lagged significantly. It is stuck in a wide trading range and is no where near break out levels. If it does approach the 370 level without looking too overbought and if the k-ratio is low enough, then I might change my overall outlook on this sector.