Sunday, February 2, 2003
Last week, I left off saying that we should expect a pickup in volatility,
and that I did not see a compelling reason to focus on the long side.
Things are beginning to change, and I would now begin to look at further
price declines as an opportunity to look for long opportunities.
____________________________________
Indicator: NYSE MEMBERS REPORT
Status: BULLISH
Comment: Notably, NYSE specialists drastically reduced their shorting activity in relation to the general public during the latest reporting period ending 1/17/03. The specialist short ratio dropped 4%, to give a current reading of 33% (meaning that specialists on the NYSE accounted for 33% of all shorting activity during that week). This places it near readings seen at past lows, and below its 5-year trading band. While this indicator gave a false signal in early June and can go much lower, the current level does suggest that public shorting has picked up to a large degree recently. During the bear market, it seems as though this indicator has made a relatively consistent series of lower highs and lower lows, suggesting that as the bear progresses, it takes a higher and higher level of public shorting before it becomes "too much" and the market snaps back. If that's the case, our current situation may become significantly more extreme before an intermediate-term low is reached. The charts below show a 4-week moving average of the specialist short ratio for two past major market lows. We can see that the 1974 low occurred with a 4-week average near 35%, while the 1982 low reached 30%. Regardless of their absolute level, both occurrences had readings that dipped below their long-term trading bands. We've seen that happen twice recently - at the July and October lows, but our current 4-week average is still hovering above the band, meaning it is not as extreme on a relative level as past lows have been. Generally, however, we have seen a high degree of public shorting over the past six months.
Bottom Line: Public shorting activity has reached a point that has coincided with market lows several times in the recent past. However, the trend of this data appears to be down during the bear market, which suggests that readings here may become significantly more extreme than what we have traditionally seen.
____________________________________
Indicator: BREADTH RATIOS
Status: BULLISH
Comment: Most of our breadth measurements are in some degree of an oversold condition. The 10-day advance/decline line is moderately oversold, but will likely relieve much of that condition this week unless we see a complete market meltdown. Over the next 5 days, the advance/decline line will drop a cumulative -4000 issues, so it's quite likely we'll see this indicator rise this week. I also like to view this indicator on a one-year stochastic basis (showing the relation of the current reading to all others over the past 52 weeks), and our current stochastic is 41 (it had reached as low as 33 this week). We normally don't see an intermediate-term low unless the stochastic troughs below 20, and especially below 10. The 10-day up volume ratio is giving a similar picture as the advance/decline line. The longer-term 21-day averages of both indicators are currently only in neutral territory, and it will take at least another week of bad breadth to get them close to oversold. The 10-day new high/new low indicator is still in neutral territory at .51. Recent lows have occurred when the indicator reached oversold territory under .20, so there is a significant amount of work to be done here in order to be comparable to past lows. The daily NYSE cumulative TICK is now oversold, and the Nasdaq indicator is slowly making its way there as well. After reaching oversold late last week, the intraday cumulative TICKS for both markets have reversed themselves and are now closer to overbought. Due to the positive days we had this week, our Down Pressure indicators are still neutral. It would take at least three days of heavy selling pressure early next week to push these indicators into positive territory.
Bottom Line: We've entered a territory here that would suggest we could see a moderate bounce, but not much more. Most of the indicators have not reached a level that normally coincides with decent intermediate-term reversals, so until we see more true extremes here, this complex is only moderately bullish at best.
____________________________________
Indicator: RYDEX RATIOS
Status: BULLISH
Comment: The asset flows in this mutual fund family have been getting quite a bit of press this week, due to the extremely high level of assets flowing into the mutual funds that profit in a market decline. In fact, the two stock index funds that are leveraged 2-to-1 on the short side (meaning they would each increase by 2% if the S&P and/or NDX declined by 1%) hit new records for asset levels this week. This is a bit deceiving, however, as the chart below illustrates:
The thick black line is the regression line (essentially an average of all of the readings). We can plainly see here that the assets in these two funds have been steadily increasing since their inception. The fact that they continue to hit new all-time highs in the face of a bear market is neither surprising nor particularly useful. However, together they have now stretched a comparable amount from their regression line as in mid-February 2002 and October 2002, both coinciding with tradable market low points. All of our other momentum indicators for these assets flows have entered bullish (to us) territory as well, on all time frames. Not only has the flow into the bearish funds (particularly the leveraged ones) been great, but the exodus from the bullish funds (once again, especially the leveraged ones) has been severe. In fact, over the past 10 days, the total amount of dollars that have moved out of the bullish funds and into the bearish is now the greatest since the leveraged funds were brought to the market in the Spring of 2000, on both straight dollar and percentage terms. The money market has stayed essentially the same during these 10 days, so there has been a $1.2 billion shift in perceptions during this January decline in these funds. While a billion dollars is nothing compared to the amount of money traded each day in the market as a whole, it does comprise nearly 30% of total assets in the stock index funds - another record. If we assume that the Rydex funds are a microcosm of the perceptions of market participants as a whole, then we can conclude that the shift in sentiment since the January peak has been unprecedented. Since we only have a few years of data to work with, I think it would be premature to jump to that conclusion, but the evidence we have seen so far is extremely promising. If we broaden our outlook a bit and look at not just the stock index funds, but all the fund groups as a whole, we once again see an unprecedented shift in psychology. Our proprietary Beta Chase Index, which measures the momentum of asset flows among the high-beta and low-beta funds that Rydex offers, hit a new record low on Friday. This shows that these market timers have shifted a huge amount of assets from the high-beta funds (which magnify stock market performance) to the low-beta or negative-beta funds (which have a low or negative correlation to stock market performance). This is more evidence that these trend-followers have pulled their heads back into their collective shell.
Bottom Line: There has been a remarkable shift in sentiment in these funds from just a couple of short weeks ago. Our measures here are at or are entering extreme territory, and taken in a vacuum suggests that we should see a snapback rally at any time. If the funds can be used as a proxy for overall investor sentiment, any rally could be sharp and extremely fast, due to a high amount of short-covering.
____________________________________
Indicator: COMMITMENTS OF TRADERS
Status: BULLISH
Comment: Two weeks ago, I pointed out the highly unusual developments that we were seeing in this data, as commercial traders were reducing their net shorts into a rising market, at the same time the small specs were reducing their net longs. Normally, a rising market prompts the exact opposite behavior from these traders. This went on right up to the peak in mid-January, and since then the unusual behavior has continued. Now the commercials are increasing their net short position while the small specs are increasing their net longs - again, contrary to the typical pattern. To be honest, I don't know what to make of the activity over the past month. There is no consistent historical pattern with which to match the current activity. If we just look at the data as we always do (in the form of the stochastics), the data remains bullish overall, as the small specs have a relatively low net long position, and the commercials are neutral to slightly bullish. The Futures Balance Matrix on a three-month and a one-year time frame remains bullish with both readings over 80.
Bottom Line: The action is this complex has gone against normal patterns for the past month. As a default, that tells me to back off reliance on this data. For what it's worth, our traditional measures continue to indicate that the data is slightly bullish overall.
____________________________________
Indicator: VOLATILITY MEASUREMENTS
Status: NEUTRAL
Comment: The excitement we saw in the traditional volatility measures (VIX and VXN) has died down a bit over the past few days, so I'm going to point out the three other measures that I keep on a regular basis and have mentioned several times in the past. These are the NYSE volume pattern, S&P average daily range and NYSE TICK range. The last time I showed charts of these measures was in the December 29th weekly commentary. At that time, I said that the indicators were suggesting that we were approaching an intermediate-term peak, so any bounce we had going forward would likely be short-lived. We got the bounce and it was indeed short-lived, so now I want to show where we are currently.
We can see from the graphs above that each measure went down to the lower end of its range in the late December/early January time period. The severe decline in most market sectors since then has served to reverse the trend in these indicators and start them on a path towards the levels they have seen at recent intermediate-term lows. However, we are not currently near a point in any of them that would suggest that low is imminent. A very consistent pattern during this bear market (and most others) is that the rallies are almost always preceded by a pickup in volatility. This is larger-than-normal moves, often in both directions, while volume picks up as well. We began to see some of that this week, which is a welcome sign that we may be entering another bottoming phase. However, as is evident from the charts above, we need to see more in order to be relatively confident that a low is being formed. If we continue to see large intraday moves with more volume, then we should find an intermediate-term low sometime in the next two weeks.
Bottom Line: We are seeing the opposite of what we saw in early January. Continued volatility with increased volume will give us more evidence that a tradable low is being formed. We're not there yet, but another one to two weeks may be enough.
____________________________________
Indicator: TRIN
Status: NEUTRAL
Comment: The 10-day TRIN for both the NYSE and Nasdaq continue to recover from the near-perfect sell signals they gave in mid-January. While the selling pressure has not been as heavy in the Nasdaq, it has been enough to bring the NYSE TRIN to an elevated level, though not yet extreme. The 10-day NYSE TRIN did come close to 1.50 during the week, but I would need to see a reading closer to 1.60 or, preferably, 1.70 before reading too much into the data here.
Bottom Line: The overbought condition of these indicators is certainly gone, but they have not yet cycled back to an area where an upside reversal would seem likely.
____________________________________
Indicator: STEM.MR MODEL
Status: NEUTRAL
Comment: Our shortest-term model did a decent job of highlighting the short-term pessimistic excesses over the past week or so. The rally on Friday was enough to alleviate the oversold condition it reached early Friday morning. At the moment, the model is firmly in neutral territory and not giving us much of a directional bias.
Bottom Line: Successive moves lower over the coming week would make it increasingly difficult for this model to enter oversold territory, but any rally would likely be quickly met with an overbought condition. Unless a larger trend change is underway, that suggests any rallies are likely to be short-lived.
____________________________________
Indicator: SEASONALITY
Status: NEUTRAL
Comment: The first few trading days of February don't have a particularly positive bias, which goes against the traditional beginning-of-month positive seasonality. Since 1950, the S&P hasn't performed much better or worse than random during the first four trading days. There may be a very slight positive bias during the first two days, but that would be stretching things. This pattern has held true over the past few years, as well, as these days have done worse than random if anything.
Bottom Line: Not much going for or against us here.
____________________________________
Indicator: SENTIMENT SURVEYS
Status: NEUTRAL
Comment: While we certainly cannot say that these surveys are exhibiting too much bearishness, they are beginning to move in that direction. The one exception, of course, and the one that gets the most press, is the Investor's Intelligence survey. I detailed why I think that survey bodes ill for the longer-term health of the market earlier this week, so I will not go over that again. On a more positive (and shorter-term) note, the other surveys are showing some constructive signs. The 4-week average of the AAII poll's bullish ratio is currently at 42%. This low of a level has been reached only 5 other times in the past decade (summer of 1993, April 1994, September 1998, July 2002 and October 2002), each one preceding a significant rally. However, each of these instances had at least two consecutive weekly readings at 42% or under, so we should expect the same this time if we're near an important low. Similar to Investor's Intelligence, the Market Vane and Consensus surveys had retained a significant amount of bullishness despite the action in the market. The latest readings began to show some improvement, and my guess is that we'll see another drop in bullishness this week, but they still remain on the lower end of neutral at best. The lesser-known lowrisk.com and German Neuer Market surveys have shown a dramatic increase in bearish sentiment over the past few weeks, and that is serving to move their 4-week moving averages close to levels that have corresponded to pessimistic extremes in the past. In fact, the 4-week average of the Neuer Market bullish ratio is currently the lowest since September 2001 and July 2002, which obviously marked lows in our own Nasdaq market. However, the German tech market has been hit considerably harder than the Nasdaq recently, which is the main reason why bearish sentiment is so prevalent there, and may be a bit less applicable to the tech market in the U.S.
Bottom Line: The sentiment surveys are showing the most constructive picture since October, but there is a significant amount of room for the bullishness in the surveys to drop before most of them could be considered extreme. And when looking for a low in a bear market, extreme is exactly what we need to see.
____________________________________
Indicator: AIM MODEL
Status: NEUTRAL
Comment: This model increased again this week, to a reading of 59.4%. The good news is that every time this model has reached and peaked near the 60% level in the past 5 years, it has corresponded to a tradable low point. The bad news is that we haven't yet formed a peak (i.e. the model is still rising), and we're not yet near readings that have been seen at recent intermediate-term lows. Each major low during this bear market has been accompanied by an AIM model reading of at least 65%. This means that we would have to see a rise in the model of nearly 6%, which is highly unlikely in the span of one week (a one-week rise of 6% only occurs about 3% of the time). So, this suggests that we would need to see yet another week of increasing bearishness at the very least, which would likely only happen if we have a down market. Since these surveys are released with a one-week delay, we may already be part-way to our goal of a 65% reading.
Bottom Line: Like most of the sentiment surveys mentioned above (and from which this model is derived), this model is suggesting that we may be within a week or two of a low-risk, tradable intermediate-term low.
____________________________________
Indicator: COMPOSITE MODEL
Status: NEUTRAL
Comment: For the first time since September 2001, July 2002 and October 2002, this model has reached the 70% level. This is a comprehensive model that takes into account the sentiment surveys, Commitments of Traders information, specialists short activity, options market activity, breadth, etc. Exceedingly high readings reflect a "perfect storm" of pessimism that normally coincide with panic sell-offs. We have not quite reached the levels of these past lows, so we cannot yet say that we are seeing a comparable level of panic across a broad spectrum of market participants. One other interesting note is that each time this model first hit the 70% level, it coincided with the time the market was entering the "waterfall stage". Entering the 70% level was NOT a positive indication - in fact, it was the complete opposite. From the time when the model first hit 70% to when the market ultimately bottomed, the average close-to-close drop was nearly 9% in the S&P 500. The average time elapsed was 8 days.
Bottom Line: This model is suggesting that we are entering a pivotal stage in the market. Recent history suggests that if we are about to find an intermediate-term low, then we are now approaching the most dangerous time to be holding long positions. While there is a high degree of pessimism, it is neither extreme enough nor broad enough to consider now to be a low-risk time to enter positions. In fact, we may now be entering the HIGHEST-risk time for the market - the time where pessimism is high enough for traders to dump their positions, but not high enough for the perception of value to be broadly accepted.
____________________________________
Indicator: STEM MODEL
Status: NEUTRAL
Comment: When this model reached its trough on January 8th, I said that it suggested we were about to go through a topping process, and not that we had necessarily seen exact high at that moment. I believe that is how this model is most effectively used, and it is telling us now that we should expect the opposite of early January - that we are about to go through a bottoming process. It doesn't necessarily mean we've seen the low for the move, only that sentiment has reached a point where the risk/reward is beginning to favor looking for longs as opposed to new shorts on an intermediate-term time frame.
Bottom Line: While the relative level of this model is neutral at the moment, the fact that it poked above its upper trading band suggests that we should begin to consider the long side once again.
____________________________________
Indicator: PUT/CALL RATIOS
Status: NEUTRAL
Comment: We have finally begun to see some heightened equity put volume (in relation to calls), which had been one of the missing pieces if looking for a market low. 8 of the past 10 days have seen an equity put/call ratio above both its lifetime and bear-market averages. This has served to push the 10-day average away from the troublesome level it reached in early January and closer to its upper trading band. However, it's only about halfway to where it "should" be if we're close to a good low point. At the same time, the OEX put/call ratio has been declining during the past two weeks, which is another positive sign (remember, I view the OEX put/call ratio in a non-contrarian light - low readings are bullish, high readings are bearish). This activity has served to move the spread between the two ratios towards the lower end of its trading band, which is a constructive sign. This spread is not extreme, but we're getting there. In the S&P 500 options, our put/call bid/ask bias ratio reached a monstrous reading of over 16.0 on Thursday, which was a good hint that we would likely see some upside over the following days. The action on Friday brought it back to a neutral reading of around 1.0.
Bottom Line: The negative connotations given by the options complex a couple of weeks ago have dissipated. Currently, the various ratios are in neutral territory, but heading in the right direction for a possible low in the coming weeks.
This weekend we have the opposite situation of a few weeks ago. Then, we had several bearish measures, some neutral and no bullish. This week, we have several bullish, some neutral and no bearish. This tells me that I want to view further price weakness as an opportunity to look for long positions, and to use caution on further short bets. There are a few things to consider here, however. One is as I mentioned above - in the past when the Composite model has reached the level it is at currently, it has coincided with the final blow-off part of the bottoming process. Although pessimism is high, it is not high enough to attract value players and bottom-fishers. That marks a dangerous black hole in the market - things look bad enough for many traders to sell, but not yet bad enough for other traders to buy. Many of our most reliable, time-tested indicators have reached levels that have signaled a low point is coming, but not levels that suggest it is imminent. With the heavy level of shorting activity evident in our current environment, I will be looking to buy - likely aggressively - another spike down. The common perception is that the October low is now the goal. That's another 10% decline, and it would correlate with the average decline as detailed in the Composite section above, but I believe we may find a low before then. I don't know at what level we will see a low-risk entry, but I think we'll know it when we see it. It is not now, but should be soon. As I said earlier this week, if the S&P can regain its breakdown point around 870, then the low may have already been put in and we will begin to see a severe amount of short-covering which takes us significantly higher. I would be reluctant to chase such a rally, but we have seen enough short selling that it may be worth it. I would much prefer to see another breakdown which serves to create a LOW-RISK entry.
- Jason Goepfert
Disclosure: no positions
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.