More Confirmation of a Low

Bottom Line: Sentiment in the bond market is as negative for bonds now as it was at the prior peak in March. It should be difficult for yields to stay this low for much longer.

Every once in a while, I touch on the current state of sentiment outside that of the U.S. equity markets. Global markets are intertwined like never before, and there are often consistent links between the commodity, currency and fixed income markets. It is on that last one that I would like to focus now.

Each day, we post several of our indicators for the bond market (see the link for Bonds under the Indicators section). A variety of data from the CFTC regarding positions in the futures markets, put/call ratios, Rydex asset flows and sentiment surveys gives us a pretty good read on how traders are feeling towards long-term interest rates. Currently, some of our readings are giving notice that the overwhelming pessimism in May has given way to acceptance of the idea that we may not be in a bond bear market after all. The best example of that attitude comes from the flow of assets into the Rydex bond fund.

We can see from the chart that assets in the fund have ballooned to their highest level in two years (actually, it’s the most amount of money in the fund since its inception in 1994). Even more strikingly, asset have increased 600% from the depths they plunged to in May, when “everyone” knew that bonds were in a for a long, hard bear market.

On the chart, we have highlighted where assets were at the March peak in bonds (point A). Now, even though Bond futures are nearly 4 points below where they were then, there is 30% more money betting on a further rise in the paper. Also, assets in the Juno fund (which profits when bonds decline) have declined some 16% in that time – the first significant decline in assets in the fund in over a year.

In opposition to these wrong-way traders, large commercial traders in bond futures are very heavily short. In fact, these traders (who hold at least 1,000 contracts) are now short to a degree unseen in six years. We would have to go back to 1998 to see them betting on a bond decline to a larger extent than now. While they were getting increasingly short while bonds rose then, historically when they have reached a short position this large, bonds have declined.

These inputs have pushed the score for our indicators to its lowest level since March, as you can see from the chart on the site (on the bond indicator page, scroll down to the link that says “Click here to see a chart of recent final scores”). If we look at the history of our indicator score, which unfortunately only goes back to August 2000, its record has been good at its extremes. The table below shows how bonds have performed over the past four years when our indicator score reached its current level or lower.

Bond Performance After Low Indicator Scores 2000 – 2004, in basis points
	5 Days Later	10 Days Later	30 Days Later	60 Days Later	90 Days Later	120 Days Later
B.P. Change	10.7	13.3	31.0	17.4	27.7	24.9
% Increases	71%	76%	90%	76%	95%	86%

From the table, we can see that 30 days later, bond yields were higher by an average of 31 basis points. For those unfamiliar with bond terminology, an increase of 31 basis points is 0.31%, equivalent to a move from 5.00% to 5.31%. Since yields move inversely to price, this also means that bond prices declined during this time. Perhaps more importantly, yields were higher 90% of the time, or 19 out of the 21 times the score reached such a low level.

If the indicator score is an accurate gauge of sentiment extremes, then when we see the opposite extremes, it should result in bond rallies. The table below shows how bonds performed after our score reached extremely high levels.

Bond Performance After High Indicator Scores 2000 – 2004, in basis points
	5 Days Later	10 Days Later	30 Days Later	60 Days Later	90 Days Later	120 Days Later
B.P. Change	(0.5)	(1.6)	(8.1)	(27.6)	(45.8)	(91.4)
% Increases	40%	36%	24%	4%	0%	0%

From the table, we see that yields declined 96% of the time (53 out of 55 days) when the indicator score was extremely high, and they were an average of nearly 28 basis points lower after 60 days. After six months, yields were lower every time, and rates had dropped by nearly one full percentage point.

The indicator score is not perfect by any means, and with a history of four years in which bonds were rising in fairly steady fashion, it has not been tested in a protracted, persistent downtrend. Still, its performance so far indicates that the current sentiment extremes should result in higher yields over the coming weeks (and possibly months), or least not appreciably and sustainably lower yields.

It’s becoming difficult to translate any potential bond move into likely outcomes for the stock market. A peak in bonds in March coincided with a bottom in stocks at the same time, while a bottom in bonds in May coincided with another bottom in stocks. The correlation between the two asset classes since the beginning of the year has been minimal and inconsistent, so it’s hard to say based on recent history what may unfold going forward. I’m no economist, but it seems as though bad economic news has already been baked into stock prices to a large degree, so should we see some positive developments on that front in the coming weeks, it should result in higher stocks and lower bonds.

Bottom Line: Despite protestations to the contrary, volume this month is actually HIGHER than it has been in recent years.

I’ve been reading more and more about how low volume has been this month, and that it is a sure sign of a lack of institutional interest in the market. The suggestion from that, then, is that any rally taking place on such tepid volume is sure to fail, as the “strong” hands are not participating.

As is so often the case, such an argument can be refuted by a simple look at recent history. Since 2001, total volume on the NYSE has remained relatively steady at an average of 1.6 billion shares daily. Looking at the distribution of that volume by month shows a clear bias towards very low volume in August, then a significant pickup in September and October. The chart below plots the average volume by month from 2001 – 2003 (the tan bars) compared to the volume so far in 2004 (the red line).

Besides the clear dip in average volume in August, what else becomes quickly apparent is that volume so far in 2004 has been coming in above average every month except for June. Perhaps most notably, volume so far in August is running 12% above average – that is the highest amount in relative terms all year except for January and May. It is true that so far this month, we have been averaging the lowest absolute amount of volume of any month yet this year, with June being a close second. However, June averaged 14% higher volume than August from 2001 – 2003, so in fact August is running well above average.

This is a very popular month for traders and money managers to take vacation. That has been a pattern for a long time, and it continues to this day. When looked at in a different light, it is evident that not only is volume so far this month not abnormally low, it is in fact higher than it usually has been during the month over the past few years.

Last week I discussed the R.O.B.O. put/call ratio and the mad scramble for put protection from the smallest of options traders. Highlighting once again the change in sentiment from these traders since the beginning of the year, this week that put/call ratio dropped only modestly, to 0.66, and they still spent 18% of their volume on put purchases. Going back to 2000, there were only 9 other weeks that showed a smaller drop in the put/call ratio while it still remained above 0.50 on a week the market performed so well. Interestingly, the two weeks when the S&P 500 showed a drop the following week, it was lower four weeks later by an average of 7.5%, but the seven times it showed a gain the following week, the S&P was higher four weeks later six times, for an average return of over 2%. So once again this admittedly small sample suggests that if the type of pessimism we see from these types of traders cannot result in a positive market, then there is a heightened chance the rally will fail.

During the week, we saw exceptionally good breadth. For three out of the five days, advancing issues on the NYSE beat declining issues by more than 3-to-1. So for the week, the up issues ratio (which is just the number of advancing issues divided by the total of advancing and declining issues) averaged 67%, a very high number that has been matched or exceeded 114 times since 1965. Out of those instances, 96 (or 84% of the total) showed the S&P 500 higher six months later, with an average return of 10.3%. This is additional confirmation of what I talked about Tuesday, when I showed other instances of extreme intraday thrusts in the TRIN.

So far, everything is playing out as it should if we have seen an intermediate-term low. We saw some true panic among those traders most likely to be incorrect at market turning points, and then we saw a buying thrust normally seen coming out of those same lows. This is longer-term behavior, and does not preclude prices from coming down to test, or even slightly exceed, the lows put in last week. But the evidence remains strong that lower short-term prices should be an opportunity to add or increase long exposure for longer-term traders.

In the short-term, as I noted in an intraday comment on Friday, the day after an options expiration has had a long-time tendency to be negative. That bias exists even going back 10 years, but recently it has been quite pronounced, especially if the week leading up to expiration was positive. Since the beginning of 2003, when the market rallied into expiration, the following Monday was higher only 2 out of 8 times, and the average return was a stiff -1%. Given some of the other extreme overbought readings we’re seeing from our short-term measures and models, we should see a breather in the advance early in the week. While I continue to think the market’s prospects look good longer-term, I would prefer to see a push lower early in the week, and our measures suggest that is more likely than a sustainable push higher.

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.