Sunday, July 20, 2003
In the last comment, I mentioned the oversold nature of our shortest-term indicators. At the time, I said that if the market cannot rally after seeing those types of readings, then that is telling us that the larger trend may be changing. On Thursday, the market was steadily weak, which is something that we hadn’t seen since the rally began in March after seeing such oversold short-term sentiment conditions. Although the market did rally weakly on Friday (continuing the expiration-Friday small-range upside bias I mentioned on Wednesday), the weakness on Thursday was unprecedented since the rally began. Something fundamental may be changing about this rally, and the mindset seems to be shifting towards selling rallies as opposed to buying the first oversold readings we get. While this is not an immediate death-knell for the market, I believe it is an important change - something not seen since February.
Many of you have asked to be kept apprised of bond market sentiment, after the large rise in yields seen last week. As an update to what I showed on Wednesday, sentiment has already become more pessimistic, as the “score” for our bond indicators has reached the highest level since early May.
On Wednesday, I mentioned that the Commitments of Traders information for 30-year bond futures was inconclusive, as the “variance” between commercial hedgers and small speculators had oscillated around zero for the past couple of weeks. That continued this week, as the negative variance from last week has switched and become positive for the most recent week (ended this past Tuesday). Commercials hedgers are now in the rare position of being more net long than are the small speculators, something which has been seen in only 13% of the weeks since March 2000. The difference is not particularly large, and we have usually seen this type of activity for several weeks in a row before bonds have bottomed in the past, so it may be premature to expect bonds to rally immediately from here. But this is a decent start, and how these groups of traders have reacted to the recent plunge in bond futures should be telling when the next report rolls out this coming Friday.
The 10-day and 21-day put/call ratios on 30-year bond futures have edged up to “oversold” territory as of Friday, after waffling for much of the week. As I said on Wednesday, put/call ratios on bonds are not particularly effective, but they’re worth at least observing.
Rydex traders in the bond funds are continuing their bearish bets, as the Juno fund (which is supposed to GAIN 1% for every 1% DROP in long-term bonds) set new all-time asset records every day this week. There is now $563 million in the Juno fund, as opposed to only $38 million in the Bond fund (which moves in lockstep with bond price). As I showed on Wednesday, this group of traders has been very aggressive in their short-side bets over the past month or so, and while they have been correct so far, it is beginning to become extreme. I have heard the argument that these fund flows should not be accurate, since portfolio managers could be using the Juno fund to hedge long-side bond positions. While that is certainly a possibility, the amount of assets in the Juno fund is very small compared to the size of the bond market, and using a mutual fund to hedge risk does not seem like a particularly efficient or cost-effective way to hedge a bond portfolio. I’m sure it happens to some degree, but from the track record of these traders, it appears as though the bulk of the money is amateur traders who speculate, rather than hedge, on price. I am comfortable with the suggestion that these traders should be faded (i.e. used as a contrary indicator) rather than followed.
Trader opinion as measured by sentiment surveys on the bond market have come down substantially over the past few weeks, and are well off the extreme bullish opinion seen in mid-May. The Market Vane bullish percentage has declined to 61%, from 87% seen in May. As I showed on Wednesday, however, this is far from the pessimistic extremes near 40% these survey respondents reached at other major bond-market lows. The Consensus bullish percentage has dropped to a low 18% (as of two weeks ago), down from the 60% level seen in May. While the balloon has certainly been popped as far as optimistic extremes go, these surveys have not deflated to a point where we could consider them to have swung 180 degrees in the other direction.
Overall, sentiment has shifted quickly in the bond market, particularly from the middle of May. However, only the Rydex traders are showing pessimism to an extreme, so while there may be a short-term bounce in bonds, from a sentiment perspective there does not seem to be a whole lot that would support a longer-term low just yet. As it relates to equities, if the positive correlation between stocks and bonds continues as it has since the equity-market low in March, then we should expect lower prices in both bonds and stocks before a more meaningful low is seen. The positive correlation between the two markets is still somewhat low at this point, so that is a relatively big “if”.
As far as equities themselves go, sentiment continues to hold at an optimistic level, despite the fact that the broader market has been relatively unchanged for over a month.
In the S&P 500 futures, commercial hedger and small speculator positions were essentially unchanged again this week. The past three weeks have shown little change here, representative of the trading-range market we’ve been in. Over the past eight weeks, while the S&P has essentially held steady, commercials have increased their net short position by over 54,500 contracts while small speculators have increased their net longs by over 76,000 contracts. While most of this occurred during the June expiration, it is not particularly comforting that this type of maneuvering has happened during basically a flat market.
The sentiment surveys, as well, have been relatively unchanged over that same period. Our AIM model continues to hover under 40%, with this week being the sixth out of the past seven showing a reading under that level. This is one of the most sustained periods of excessive bullishness seen over the past decade, as the following chart demonstrates. The chart shows the number of consecutive weeks the model gave readings of 41% or less (I am fudging a bit here and using 41% instead of 40% since two weeks ago the model gave a reading of 40.9%). The red horizontal line highlights our current scenario, with seven straight weeks.
We can see that this amount of sustained bullishness has been seen only three other times during this period. We saw 9 straight weeks ending in February 1997, which lead to a fairly steep correction into April. Then there was 12 consecutive weeks ending in August 1997, which lead to a choppy move higher for the next six months before the market broke higher. Lastly, there were 12 weeks ending in April 1998 which lead to choppy trading for a month before a final short-live spurt higher into July, then the eventual crash into September of that year. After each of these instances reached our current number of 7 straight weeks, the upside in the S&P was limited to a maximum of about 6% before the streak finally ended.
The put/call ratios are showing an interesting confluence this week. In an abnormal development, both the equity-only put/call ratio (with and without QQQ options) and the OEX put/call ratio are near the lower ends of their ranges on a 10-day and 21-day moving average basis. Typically, we’ll see the equity p/c ratio decline while the OEX p/c ratio rises, and that is most often observed near market peaks when it becomes extreme. Or, we’ll see the equity p/c rise while the OEX p/c ratio declines, and that is most often seen near market lows. This relationship is not perfect, as the correlation between the two 10-day averages since 1997 is -0.31, which can be considered only mildly negative. But at the extremes, the two most often diverge strongly which can be clearly seen near the mid-June highs. About the only recent precedent I can find for this type of behavior is mid-April of this year, when both ratios were declining rapidly toward the lower ends of their trading ranges. Of course, that instance resulted in the move higher into June. Going back over the past six years of data, I could not find a consistent relationship between this type of activity and future market performance. Sometimes it lead to market lows, sometimes market peaks, and sometimes just random bouncing around. While on the surface such preference for calls over puts across these trader groups seems as though it should be bearish, I cannot find any evidence for or against that one way or the other – to me, it appears neutral. The bearish implications from a low equity put/call ratio are balanced out by the positive implications from a low OEX put/call ratio. I would avoid reading too much into these figures for the time being.
A little over a month ago, I began suggesting that a trading range was likely for the foreseeable future, so an emphasis on oscillator-type indicators would pay off more than trend-continuation type signals. That has worked relatively well, as oversold signals have been getting bought while overbought has been getting sold. At some point, obviously, we will break the range and should travel some distance once it is broken. From my perspective, it still appears as though a move towards the recent highs will get sold, as momentum has waned and the character of the market has changed (e.g. we were not able to rally strongly at the first signs of an oversold condition). My preference remains with shorting rallies unless we exceed (and hold) above the 1015 level on the S&P 500.
- Jason Goepfert
Disclosure: long OEX puts, long SPX puts
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.
© 2003 Sundial Capital Research, Inc. All Rights Reserved.