BI-WEEKLY MARKET RECAP and TECHNICAL PERSPECTIVE

By: Larry Katz

September 4, 2001

Capsule

Over the very short-term we see enough evidence to support the idea that a rally is close at hand. Any rally that does emerge from current levels, however, should be viewed as only short-term in nature and once complete the medium-term downtrend should begin to re-assert itself. On a long-term basis, the evidence is overwhelming that the bear market is far from over. The bonds are very stretched on a short-term basis and could correct at any time. The medium-term picture is still OK but it too is getting a bit long in the tooth. Sentiment on gold is excessively negative. The XAU has turned bearish for the short-term and remains so for the medium-term

Elliott Wave and Fibonacci

When we wrote our long-term forecast in early 1999 we had two main themes. The first was that the 1550 area on the S&P was going to be a very important resistance point and a major price objective for the completion of intermediate wave 3 from 1982. The second was that since wave 2, which was the 10987 crash, was a simple pattern in both time and price, wave 4 following Elliott’s rule of alternation would be complex in either time or price, or perhaps both. This has certainly been the case so far as the S&P made its price high in March of 2000 and most likely its orthodox peak of wave 3 in September of last year. The pattern from either of those peaks, and we are favoring September of last year as the orthodox peak has been a lot more complex and time consuming than wave 2. This again supports the idea that our preferred count is correct and that the post 2000 decline is indeed a fourth wave. The decline from September of 2000 into the March low was in our view a corrective structure and is either a double three or a simple a-b-c with wave c unfolding from early February to late March. The rally from March to May retraced far too much of the January-March decline to be related to that pattern and as such we are of the view that it was related to the entire post September 2000 pattern. In fact, the May 22 high slightly exceeded a 50% retracement of the entire decline. Yes, we know that 50% is not a Fibonacci level but it an important zone of resistance nonetheless. The decline from May has exceeded a .618 retracement of the March-May rally, which by the way was a clear three-wave pattern basis the weekly charts. The S&P did stop last week right close to a .786 retracement of the rally. This is not a Fibonacci number by the books but it is important as it represents the square root of .618. We had thought it possible that the S&P was entering a third or “c” wave from the May high at the August 2 peak. However, the S&P moved above the previous weeks high last week and that basis the weekly charts locked in the decline from August 2 to August 22 as a three no matter how you slice it. Unless there is some sort of diagonal triangle here that we cannot yet see it is virtually impossible to count the pattern as impulsive, and that in turn eliminates the possibility that the S&P did trace out a five from May 22 to July 11. Therefore we are counting the decline from May 22 to July 24 as a three and the July 24-August 2 rally as an X wave. The post August 2 decline is unfolding as a simple a-b-c and is best viewed as a second three from May 22. It is possible that the decline from May 22 is a “b” wave from March but in our view if this is the case we need to hold near current levels and begin to rally in very impulsive fashion. Given the medium-term technical backdrop we do not think this is the likely case and instead are counting the post May decline as the beginning of a second three from the September 2000 peak. The decline from August 2 into last weeks low has modestly exceeded being .618 the May- July decline. Moreover, the decline from last weeks high into Thursday has exceeded the .618 relationship with the decline from August 2 to August 22 or the “a’ wave of the second three. The decline from last Monday to Thursday can be counted as a five on the hourly chart and that does allow for a rally. However, the daily chart remains down so it is possible to see the pattern extend. However, the fact that we do have a five does allow for the S&P to rally over the near-term. While it is possible that the low last week can be counted as a completed “b” wave from March, internal Fibonacci relationships argue against that count. As such we are counting last weeks five-wave decline as wave 1 of “C” from August 2. There are two separate price relationships within the post May structure that point to the 1068-1078 zone as important. First and most important is that the post August 2 decline would be equal to the May-July decline at 1078. Secondly, the decline from August 27 (wave “c” of the second three) would be 1.618 wave “a” (August 2-August 22) at 1068. There is some intervening support at 113 where wave “c” from August 2 would be equal to wave “a” but from the looks of things a move to or slightly through the March low looks likely at this point. The decline from last weeks high on the DJIA can be counted as a very clean five-wave pattern on the hourly chart but has yet to be confirmed as such on the daily chart. The latter does allow for the possibility that the decline can extend. This decline may be counted as a fifth wave from May 22 but only if we have a very unorthodox triangle ending on August 27 going back to the June 26 low.  However, in our view while this is possible under the wave theory it is not the best of counts. As such we are going with the idea that the DJIA has just entered a “c’ wave from July 19 and that the post July 19 decline is a second three from May 22. It is possible that at Thursdays low the pattern is complete and it is also possible that the entire post May decline is over and can be counted as a “b” wave from March. The low on Thursday only slightly exceeded a .618 retracement of the March-May rally and at last weeks low the post July 19 decline stopped very close to being .618 the length of the May 22-July 11 decline. The fact that we have five down hourly from last weeks high does set the stage for a rally.. At this point we are going with the idea that last weeks decline was the first wave of a larger five-wave “c” wave from July 19. And that the post July 19 decline is a second three from May 22. The alternate count is that the decline from July 19 to last weeks low was only wave “a” of the second three. This would allow for a deeper rally now to be followed by a larger “c” wave decline. How we rally, if indeed we do rally will be important to the medium-term count. The NDX is keeping alive the possibility that the decline from May 22 is a fifth wave from the March 2000 peak. If that is the case we are still likely in wave 3 of iii, which is still subdividing. The decline last week moved well below the .786 retracement of the April-May rally with the NDX less than 90 points from the April low. So while it is still possible that the decline from May 22 is indeed a “b” wave from April it is getting a little late for that count to be correct. The hourly chart from last weeks high to low can be counted as a five and that does allow

For a rally over the near-term. At this point we are approaching any rally as being directly related to the decline from august 27-August 30 and we are expecting to see the April lows broken. Support S&P; 1120-1124, 1111-1113, 1068-1078, DJIA 9820-9850, 9500-9550, NDX; 1400-1402, 1380-1384. Resistance: S&P; 1148-1150, 1162-1165, DJIA; 10,088-10,100, 10,222-10,235, NDX; 1498-1501, 1532-1537. Trend changes for the next two weeks are as follows, September 4-5**, September 8**, September 13-14. *denotes potentially important.

Bonds

The bonds have continued to rally benefiting nicely from weakness in the equity market. The bonds have not yet moved above the March high so it is still possible that the rally from May 17 is a very deep second wave from Marc. However, it is more likely that the rally from May is the final leg of the rally from January 2000 and a move above the March high is a distinct possibility. The daily range oscillators are beginning to turn down from overbought levels and some are showing minor negative divergences. The daily trend oscillators are still positive but are also beginning to diverge and on balance short-term momentum is neutral to weak. The weekly range oscillators are heading towards overbought levels but are still OK. The weekly trend oscillators are positive and medium-term momentum is positive.. Market Vane reported 64% bulls and Consensus Inc 585. These are negative and a warning but are not yet excessive. The commitment of traders report shows that commercials hedgers are net short but here too it is not at extreme levels. On the whole, sentiment is negative but not yet excessive. Support: 104 12/32 +/- 5/32, 103 16/32 +/- 7/32. Resistance: 106 10/32-106/20/32, 107 4/32. Please note that all support resistance levels are based on a constant bond chart and not any one futures contract. Currently, the constant bond contract is 3/32 ahead of the December futures. The bottom line is that the short-term picture is beginning to weaken and a correction looks close at hand. The medium-term picture is still positive so any weakness should be short-term and followed by further gains. We are neutral short-term and remain bullish medium-term.

XAU

The rally from the July 2 low has retraced about 50% of the May-July decline. The rally is a three-wave structure on the daily and weekly chart. So to is the May-July decline. That said, the rally decline from May to July is most likely the “a” wave of a larger pattern. The rally from July 2 is either all of or only the “a’ wave of a “b” wave. However, the fact that it is corrective argues strongly that the July 2 low should be broken before the minimum for the completion of the Post May decline is in place. Whether that leads to a break of the October 2000 low is another question and one we cannot answer at this time. Short-term momentum is rolling over from overbought levels ands has turned negative. Medium-term momentum is improving but is still negative. Some sentiment indicators have turned downright awful. Market vane and Consensus Inc are in the high 50% to high 60% area. The Rydex precious metals assets have been in excess of 70 million for days on end. This is the level seen near most important tops of the past several years, Moreover, the commitment of traders report shows a huge net short position by the commercial hedgers and a significant net long position on the part of the small speculator. Support: 54.60-55, 5350-53.70. Resistance: 58.90, 61-61.24 The bottom-line is that both the short and medium-term are negative. Long-term we are neutral.

Indicator review

August came in with a whimper and went out with a bang. The DJIA lost over 570 points or 5.4%. The S&P lost 78 points or 6.4% and even the broad based NYSE Composite was hit hard, losing over30 points or 4.85%. The last week of the month was particularly weak with all three averages recording very strong losses and closing lower four of the fiver trading days. The averages broke important medium-term chart support completing topping patterns and closed at their lowest levels sine early April. So much for the summer rally. The S&P remains well below its steeply declining 200-day moving average. The DJIA broke well below its 200-day moving average, which is now beginning to show a negative trend. The NYSE did likewise and its 200-day average is starting to decline instead of going flat. Price/volume relationships remains quite negative and in fact became more so  last week with volume continuing to decline on rallies and expand on declines. However, we have not seen anything close to what could be viewed as climactic behavior from volume. Breadth started to show some signs of weakening last week but on balance has been relatively decent and the A/D lines, both daily and weekly have held up far better than the averages. The high/low statistics have been unusual. We have seen a good number of new highs on a daily basis, although that number has been dominated by closed end funds and preferred issues both of which are driven more by the bond market. When the daily stats are adjusted for those issues they are not nearly as strong. We have just begun to see a modest rise in the new lows, and the high/low indicators have turned down and issued sell signals. The weekly numbers are showing a similar pattern of behavior. The Russell 2000 lost 17 points or 3.55% in august. It did hold up better than the big cap averages but nonetheless did break important chart support. We are going to remain neutral on the short-term but with a negative bias. Our medium-term negative rating remains firmly in place. The Value-line lost 47 points or 3.83% and like the Russell did hold up better than the big cap averages. However, also like the Russell it too broke important medium-term chart support completing a negative pattern. We are neutral short-term but with a clear negative bias. The medium-term remains bearish. The NASDAQ Composite lost 222 points or 10.9%. The NASDAQ 100 lost 214 points or 12.7%. They are neutral short-term but also with a negative bias. The medium-term remains negative. The DJTA lost 90 points or close to 3% last month. It did hold up a bit better than the DJIA but was still weak. The short-term is neutral but beginning to weaken. The medium-term is bearish. The DJUA and UTY did close the month lower but only by a very minor amount. They also managed to hold above their July low. We are neutral short-term but that is beginning to weaken. The medium-term and especially the long-term remain bearish.

Momentum  

The breadth oscillator is still only neutral but is showing sings of breaking down. The volume oscillator is on the oversold side of the equation and is a modest plus.  The 3-day oscillator reached minor oversold readings late last week and is back to neutral. The McClellan oscillator is neutral but did break down out of a near two-month range. It is weak and the summation index has turned down and beginning to gap. The 10-day and open 10-day Arms are oversold but are not nearly as extreme as they were two weeks ago. The 5-day Arms is back to oversold after moving to neutral and the 21-day Arms is also oversold. The new 10-day Arms is neutral but shoeing some improvement. However, it is still on a sell signal from August 15. The daily range oscillators are near oversold but are in very weak patterns and are negative. The daily trend oscillators are negative. The weekly breadth oscillators are neutral but weak and just beginning to roll over. The weekly range oscillators are neutral but getting closer to oversold. The weekly trend oscillators are still very negative. The inside score of the technical barometer did improve but only slightly so and is only neutral. The outside score is also neutral and has a long way to go before turning even remotely positive.

Sentiment 

The sentiment model is still borderline negative and has a long way to go before turning positive. Investors Intelligence has seen the percentage of bulls drop a bit and that is improving but the number of bears is still extremely low and this key indicator has a long way to go before even hitting neutral levels. Market Vane and Consensus Inc have improved a bit but are still only neutral. We have also seen some improvement in the American Association of Individual Investors (AAII) survey but this too has a ways to go before getting close to positive. NYSE members were on the sell side for the third straight week and quite heavily we might add. This key indicator if smart money has moved to neutral from bullish. The Insider sell/buy ratio has eased but is still on the negative side of the equation and has a way to go before hitting neutral. The commitment of traders report is still extremely bearish as commercial hedgers remain heavily net short while the small speculator is still heavily net long. The 10-day moving average of the CBOE put to call ratio is OK but is still only neutral. The OEX ratio is neutral but closer to overbought. The Rydex ratios are in decent shape and at levels that could support a short-term rally. However, they are not yet at levels that could support a medium-term low and are far from levels that would suggests an important major low.  The VIX has moved up and is at levels consistent with short-term rallies but it is not close to where it was at the March-April low. The VXN is not even at levels consistent with short-term low but it has moved up and is not extremely negative either.

Summary and Conclusion

In an interview this past weekend we were asked what it would take to turn us bullish again. We are going to answer that question in a round about way. Back in the latter stages of what is now widely accepted as the culminating phase of the technology bubble, we were on a daily basis exposed to such comments as technology stocks are immune to the business cycle. Technology stocks are immune to higher interest rates. This time its different. We are in a new paradigm. The old indicators do not work anymore. We heard these comments so often we almost began to believe that they were right. This publication, however, spent many an issue suggesting the opposite. That it is not different this time it only feels like it is. It is next to impossible to know when a bubble will ultimately burst especially when you are in the eye of it. However, they all do burst and when they do the consequences are the same. The peak of the bubble is a top that is not visited again for years, perhaps even decades. A prime example is the Japanese market. The Nikkei made its all time high in late 1989. It is 11 ½ years latter and it is still down over 29,000 points or nearly 70% and at 16 year lows. In 1929 the DJIA completed a top of epic proportions and it took 25 years to get even. This in that round about way brings us back to the current market and what it might take to turn us bullish again. In bull markets the market shrugs off bad news and even rallies on it while good news is greeted with a rash of buying.  In bear markets good ness produces little or no rally and bad news sends stocks sharply lower. Does the latter sound familiar?  The first thing we would want to see is a change in the markets reaction to news and its perception of news. As we have often said, it is not the news that matters but how the market reacts to the news that is important. And right now it is acting in typical bear market fashion. However, more importantly than that, there is a certain psychology that has accompanied nearly every important bear market low. We have not at this point come close. And yes while it may b different this time it most likely will not. In our last report we discussed a few longer-term indicators that need to show significant improvement before we could even remotely begin to start looking for a bottom of importance. Those indicators such as the Investors Intelligence percentage of bears is not even close. At the final low we would expect to see this indicator at least in the high 40% zone but more likely in excess of 50%.  At the March low it was over 42% so as we can see this indicator still has a long way to go before even coming close to levels that could support a medium-term low let alone signal that we are getting closer to the end of a bear market. Some of our short-term indicators have improved mostly basis hourly data but some such as the 3-day oscillator are also OK. The short-term wave structure shows what can be counted as a completed pattern on the hourly chart from last weeks high. The VIX has moved up to short-term rally levels, however, the VXN is a bit disappointing even short-term as it is not yet close to levels seen at other short-term lows. The Rydex ratios are OK and the CBOE put to call ratio moved higher late last week in response to the sharp drop in prices. All of this adds up to the strong likelihood that we could see a sharp short-term rally get underway at anytime and last for a few days. We are also supported by the fact that the market for the past 8 or more months has been able to rally into the first week of a new month almost religiously making a top in the first week of the new month. This pattern actually stretches back to September of last year. This occurred regardless of whether the market was able to rally late the previous month or not. While we do see the potential for a rally to begin as early as today, we do not see that this rally has the potential to amount to very much other than over the very near-term and at this point it is our view that the rally, if it does occur will most likely be related to the decline from August 24 and not much else. As such we are going to remain in the neutral camp for the short-term. Medium-term we have seen some improvement from some areas but overall the picture is still quite negative. The technical barometer for one is only at neutral levels and lets not forget that we did get an extreme –10 reading in late April. In addition, the sentiment composite is only at +8, which is still borderline bearish. At the March-April low for example it moved to a +14. Once the short-term rally is complete we would expect the medium-term downtrend to re assert it self and see lower prices ahead. We remain bearish on the medium-term and also very much bearish on the long-term.

QQQ traders are holding a 50% long position from 36.30. Keep your stops at 34.30. Rydex switchers are holding a 10% position in Ursa. Make sure to call the Noon Pacific hotline for any changes.  

In spite of the decline over the past several weeks the sentiment model has shown no improvement and remains at borderline negative levels. Note the spike at the March low and also at the October 199 low.  At the 1998 low the model moved to bullish levels and we suspect we are going to have to see something of that nature before we could consider the idea of an important low.

 A lot has been said about the 10-day and open 10-day Arms of late. And it is difficult to argue with its historical significance. However, we also look at and report the new 10-day arms developed by Peter Eliades. While the other have been bullish the new Arms has clearly not been and its last sell signal on August 15 looks real good. It has improved but is not even close to where it was at good trading lows such as March 2000 October of last year and even at the March low of this year.

The VIX is back to levels seen near some minor trading lows over the past few months and that could support a rally. However, it is not close to where it was at the March low nor even close to where it was at good trading lows seen in October and December of last year.