Thursday, October 23, 2003 11:30 PM EST
As I write this, the S&P futures are trading around 1021. This general area, within a point or two, brings together three of the most widely-watched and obvious support zones for short-term traders. Namely, it closes the gap from October 3rd, it touches the 50-day moving average, and it is a 50% retracement of the rally from September 30th. As you know, I try to leave technical factors out of these comments, but this kind of confluence is so obvious that it will be used by a great number of traders as an inflection point. Meaning, traders will be using the 1020 area as a “line in the sand” – bullish above, bearish below.
From a sentiment perspective, I don’t have a whole lot to support either side at the moment, especially short-term. As I mentioned this past weekend, we are still mired in what is historically the weakest time of the month of October, though that switches to a positive bias beginning early next week. Most of our short-term indicators are now in neutral territory, and in fact the “score” for those indicators is now exactly neutral at 50%.
Each day, more and more of our intermediate-term indicators are relieving their over-enthusiastic extremes. This does not mean that they are becoming positive for the market, they are just becoming less negative. However, in the context of a longer-term uptrend, oversold is a rare condition and more often than not all it takes is simply some relief of overbought conditions before the trend can resume.
CONCLUSION
Tonight’s comment is “lighter” than usual, as there is just not much to say that differs from what I have gone over during the past week. As I said on Tuesday, my preference has been to the short side of the market, since we were nearer the top end of the recent range and each time it looked as though new highs were a foregone conclusion, traders would pile in and we would slip right back into the range. On the flip side, each time it looked like the market was finally going to crack and put in an intermediate-term high, buying pressure would come in and we were taken up to or through previous highs. Each time we have sold off by a few percent, the shorts piled in and the losses were reversed.
We are beginning to see some of that already. Just a couple of percent off the highs, the put/call ratios are perking up (such as the CBOE equity p/c minus QQQ options), and the Rydex fund flows have been getting more skewed to the bear side. The Rydex Beta Chase Index and RSI Spread are already entering positive (for the market) territory, though there is still a lot of room for the longer-term measures there to work off some of the excesses we’ve seen since the beginning of October. Nothing that I follow is at an oversold extreme, even on the shortest of time frames. That would likely change with some sustained weakness tomorrow, but so far I don’t see anything from a sentiment perspective to get excited about. Like I said above, it’s likely we will have some type of bounce off that 1020 level simply because so many others are watching it, but it would be a real stretch to suggest a bounce should occur due to overwhelming pessimism. It still looks as though bounces that relieve any short-term oversold conditions are meant to be shorted, at least until we see more pessimistic readings like early August and late September.
On a few side notes, I want to mention that due to numerous requests, I’ve now added the Rydex Enthusiasm Index and the VXN Fear Premium to the regular site updates, and they can be found in the Indicator section of the site. Also, for those of you who know what charts you’re looking for, I have added a “quick pick” section where you can choose the chart from a drop-down list and have it appear immediately below. This saves you a lot of clicking and scrolling and opening and closing windows. Lastly, I have begun updating the intraday charts several times daily, besides the regular 11:00am EST and 2:30pm EST updates. Now those charts are updated usually at least six times daily.
- Jason Goepfert
Disclosure: long OEX 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.
Tuesday, October 21, 2003 9:35 PM EST
Margin debt in customer accounts at NYSE firms jumped 4% in September from what was seen in August. At the same time, free credit balances increased just a tad. In the comment on October 8th , I showed a chart that tracked the “available cash” or “net worth” of these customer accounts. The available cash was simply the free credit balances (assets) minus margin debit balances (liabilities). The changes for September decreased the net worth for these customers by about $5.6 Billion, so it currently stands at negative $67.2 Billion - more than double what it was in September 2002. This is the worst position the customers have been in since June 2001, but it is a far cry from the worst-ever at negative $210 Billion in February 2000.
I’ve said before that the increasing use of margin can actually be a GOOD thing for the market, as it provides a source of fuel for further gains. When it becomes extreme, that’s when the trouble sets in. The difficulty lies in determining what, exactly, constitutes an extreme. For this data, I prefer to look at year-over-year changes, and in that regard we are about as extreme as we were in August 2000, but still far from the ultimate extreme seen in the Spring of that year. Overall, my take is that this data is still relatively neutral for the market’s potential going forward – it is good to see margin accounts being used again, but the velocity with which it is being pursued may be a little too much, too soon.
SHORT INTEREST
Short interest in NYSE securities increased a little over 1% in October from what was seen in September. On the surface, this seems like a positive, as it shows that traders were shorting shares into a generally rising market. However, when we adjust the figures for volume patterns and the seasonality seen in short interest itself (both vital adjustments, in my opinion), then the short interest ratio actually dropped by a small amount. As you can see from the site, this short interest ratio continues to be the lowest seen in over 10 years, and I would regard it as a very mild negative for stocks going forward.
NEW HIGHS
One aspect of the recent rally that is troubling is the shrinking number of new highs. This shows that fewer and fewer of the leading issues are participating in further rally attempts. A winnowing down of the leader board is oftentimes a preliminary signal that buyers are becoming exhausted. Most of you have likely heard and seen this argument several times, so I’m not going to belabor the point. I do, however, want to show why this type of analysis can be useful. The two charts below show the number of stocks hitting new 52-week highs on the NYSE for 1998 on the left and for our current period on the right. The number of new highs is shown by the tan columns and the S&P is the black line.
|
|
|
If we look at 1998 first, we can see that as the market rallied strongly throughout the Spring (point #1 on the left chart), it was accompanied by an ever-increasing number of new highs. By March, there was a final climactic push to 372 new highs (#2) which gradually subsided as the market formed a trading range over the next few months. The final push higher in the S&P later that summer was accompanied by only 180 new highs (#3) even though the market had pushed significantly above the high set in the Spring.
Currently, we are seeing a somewhat similar situation. The rally in the Spring (#1 on the right chart) was met with a ramping up in the number of new highs, which culminated in a final push to 592 new highs in June (#2). Since that time, there have been two distinct thrusts to new highs in the S&P, both of which have come on decreasing numbers of new highs: first 508 in September (#3), then 498 in early October (#4). Today I show 227 new highs, even though we’re only 4 points from a new closing high (though it’s more constructive to look at the figures when we actually make a new high in the broader market).
This is certainly not as dramatic as what was seen in 1998, but I want to point out the usefulness of tracking this type of information. Some of you will probably have different figures for the number of new highs, and some of you will also point out the effects of closed-end funds, non-operating companies, decimalization, etc. on the figures. Those are valid concerns and if I was trying to suggest that the market should fall now because of these divergences, then I think they should be accounted for. I am not suggesting a decline (based on this information alone) – I am simply trying to give a heads-up as to a potential issue should we continue to rally.
VOLATILITY
The VIX has received its usual fair amount of attention as it continues to hit new multi-year lows (both old and new formulas). The new VIX is now the lowest it has been on an absolute basis since August 31, 2000. I think viewing the VIX on an absolute basis is not only misleading, but it’s dangerous. I much prefer relative measures, such as looking at it compared to actual historical volatility (e.g. the VIX Fear Premium), or compared to other VIX readings in the recent past. One of my preferred measures is looking at how far the 10-day moving average of the VIX is stretched from its 252-day moving average. This gives us a fairly good approximation of extremes, and enables us to more accurately compare our current situation to that of past periods.
In that regard, currently the 10-day moving average of the VIX is about 27% below its one-year average. This places what we are seeing now in the bottom 5% of days since 1990 (again, using the new formula). There are really only four other periods that can be compared to the current one, each of which is shown below. This series of charts shows each period where the 10-day average of the VIX was stretched more than 25% below its 252-day average. The red boxes highlight the period from when the indicator first turned up from its lowest point to when it went back to “normal”, meaning when the 10-day average became approximately equal to the 252-day average. For the purposes of these charts, I used the old VIX calculation since we have more history for that. In checking the old formula against the new one, there was really no difference as far as this indicator is concerned.
We can see that in 1998, the S&P dipped as the indicator first turned up, then rallied strongly before declining again, to finally end up about where it began. In 1991, the S&P basically went nowhere for six months while this condition was worked off. In 1999, the S&P dropped hard and immediately, with the VIX quickly reversing its oversold condition. The same thing happened in 2002, as the bear market resumed in full force.
Now let’s look at the current period:
Clearly, this is different from all the others. We have already gone nearly six months with barely a correction while the indicator has stayed beneath -20% for essentially the entire time. While it’s tempting to say “this time is different”, so the VIX could stay stretched for a long time to come, I would feel more comfortable suggesting that the current period of low volatility will revert to the mean by increasing. As you know, increasing volatility is almost always accompanied by declining stock prices, not rising ones. Obviously, the argument for rising volatility and declining stock prices could have been made for six months now when the indicator above first became extreme and began to reverse, all to no avail.
CONCLUSION
I can point out minor flaws with this market for pages and pages. Together, they paint a compelling picture that stresses the short side is the place to be. Taken in isolation, however, each is little more than a minor quibble. Each time it looks like we are about to break lower, buying pressure comes in and we are taken up to or through new highs. When we do hit new highs, however, the bandwagons become overcrowded and we slip right back into the range. When we sell off a few percent, and shorts pile in like in late September, and the losses are reversed. This is a pattern that has repeated consistently for the past five months.
This weekend, I mentioned that I would be inclined to look for a bounce of 15-20 S&P points, if the weakness (and short-side bandwagon-jumping) that we saw on Friday continued for a day or two. We never got that chance, and have already rallied about 15 points. My assumption going forward is that the pattern that we have seen will continue. We may break marginally to new highs, but won’t get much further than that before a correction back into the range sets in. If we drop 3%-5% from the highs, then the short side will become too crowded and they will be punished accordingly. Since we’re nearer the upper end of the range, my preference continues to be with the short side, and that would likely only become stronger should we poke above the recent highs. Should we instead decline a few percent from here, I’m confident that most of our gauges will show pessimism working up again, and at that time I think the long side would be a safer bet.
- Jason Goepfert
Disclosure: long OEX 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.
