A Link from Metals to Bonds to Stocks

Bottom Line: COT analysis for precious metals suggests a further decline in the XAU is likely; a decline in the XAU should lead to a rally in bonds; a rally in bonds should lead to a rally in stocks.

I don’t often touch on sentiment outside the U.S. equities market, but with so much volatility occurring in other sectors, such as bonds and commodities, I have been getting an increasing number of requests to update sentiment measures on those markets as well. I took a look at bonds last week, but the pounding being taken by precious metals recently is causing them to take bonds’ place as the most-requested market upon which to focus.

By far, the most requests I’ve received are for the gold and silver markets. The moves in precious metals have been remarkable recently, with gold futures down nearly 10% and silver futures down nearly 30% since the beginning of April. The interest here is not for entertainment purposes only - I showed last week the correlations between the bond and stock markets, but there is also a correlation between precious metals and bonds.

The scatter-plot below shows the correlation between one-month changes in the XAU (Gold and Silver Index) and future one-month changes in 30-Year Treasury Bonds from 1986 through the present.

The correlation is -0.18, and with 755 data points, the probability that this is due to chance alone is basically zero. So we know that in a broad sense, a rise in the XAU one month tends to lead to a decline in bonds the next month and vice-versa. Since we know that bonds and stocks have a very distinct correlation, this additional link between the XAU and bonds may help with clarifying the outlook for stocks as well, and not just for those with an interest in precious metals.

We follow the Commitments of Traders data for stocks and bonds on the site, but the information is available for a whole slew of commodities, including precious metals. Often, the data is more useful in these other markets than it is for equities or even bonds. The chart below is a composite of the positions for gold and silver futures, adding the net positions for each major trader group together for both markets. It shows the XAU (top, in black), the net positions of small speculators (in red), the net positions of large speculators (in blue) and finally the net positions of commercial hedgers (bottom, in green). The dotted vertical lines showing the division between the years is scrunched up on the left side of the chart because the CFTC released this data bi-weekly up until 1992, so that data is compressed compared to post-1992.

As of 04/12/04, commercial hedgers in gold and silver futures (combined) were short a record number of contracts, large speculators were long a record number of contracts, and small speculators were close to showing a record net long position. As is true with most commodities, when small and large specs are extremely long, the market tends to decline soon afterward, and commercial traders just happen to almost always be very net short at the time. With each of the groups at or near historic extremes, the picture could not have been more bearish for precious metals.

Another way of looking at this is to combine all the positions into one “master” look at all the positions in both gold and silver. This measure subtracts all small and large speculator net positions from commercial traders in both metals and gives us one easy look at whether these traders’ positions are likely bullish or bearish for the metals, and the XAU by extension. When this indicator is high, it means commercial traders are long and/or speculators are short, and it should be bullish for the metals. When the indicator is low, it means commercials are net short and/or speculators are net long, and it should be bearish for the metals going forward.

Last week this composite indicator was so negative that it was nearly twice as extreme as the previous lows in the mid-1990’s. To see if this indicator has had any predictive value, we can look at how the XAU performed after the indicator reached an extreme. For these purposes, we will define “extreme” as simply either the lowest or highest level it had seen in the past year. The table below outlines how the XAU performed the given number of months after the “master” net futures positions indicator reached a new yearly low (“0”) or a new yearly high (“100”).

XAU Performance After Net Position Extremes 1986 - 2004
	When “Master” is 0			When “Master” is 100
	1 Month Later	3 Months Later	6 Months Later	1 Month Later	3 Months Later	6 Months Later
Avg Return	-1.3%	0.1%	-1.7%	5.8%	6.7%	10.6%
% Positive	45%	46%	45%	66%	71%	68%

Since this composite record of positions was at a new all-time low two weeks ago, it suggests that the XAU may have difficulty making significant headway over the intermediate-term. If the historical correlation between the XAU and bonds holds, then a decline in the XAU should lead to a rise in bonds. And since stocks and bonds have had a very strong positive correlation since the blowout employment report on April 2^nd, it further suggests that stocks also should rise.

Bottom Line: The “score” for our bond sentiment indicators is at a level matched only a few other times in the past four years – each of which lead to significantly lower rates within 30 days.

As an update to the bond sentiment data I posted last week, we have become slightly more extreme in terms of negative sentiment. The chart below is the composite indicator score that is posted to the site each day, and it shows that at a current reading of 1.33, we have seen similar bouts of pessimism only a few other times in the past four years.

Looking at the other instances, 10 days after the score first reached the current level, 30-year T-Bond yields were lower every time, with an average decline of 1.4% (since price and yield move in opposite directions, a decline in yield means bond prices rose). After 30 days, yields were still lower every time, by an average of 5.1%; after 120 days, once again yields were lower ever time, and the average drop was a whopping 13.5%. Obviously, to project a similar type of reaction for the current occurrence would mean that we have to assume that the long-term trend in bonds is still up.

Most of our breadth measurements (up issues, up volume and cumulative TICKs) are just coming out of deeply oversold territory, and suggest that Thursday’s action was only the beginning of a longer move up. While there is certainly some strong technical resistance immediately above the major indices, I don’t see much from a sentiment perspective that supports those resistance levels, suggesting that we should be able to poke higher and test new highs. For the past few weeks, I have been using the 1120ish level on the S&P 500 cash index as an initial warning sign that a deeper correction may be in store, and so far that index is holding pretty much exactly as it should. I continue to find a lack of strong reasons to bet against this uptrend and I continue to favor trading from the long side.

This disclosure is not intended as trading advice in any form. It is meant as a note to subscribers that the author may have a position directly affected by the market outlook reflected in the commentary. Although the author takes great pains to remain objective in any commentaries, it is only fair that readers should know that the author may have taken positions in accordance with his market outlook. Positions can and do change at any time, without notice to the reader.