BI-WEEKLY MARKET RECAP and TECHNICAL PERSPECTIVE

By: Larry Katz

December 10, 2001

Capsule

A number of important divergences are developing from a number of momentum indicators and key sentiment indicators are weakening as well. The short-term is difficult but looking sloppy and weak and a decline could get underway at any time. We remain cautious. The medium-term picture is deteriorating rapidly as a number of indicators have weakened. We are close to a top but do not feel we are quite there so higher prices as the topping process progresses is the likely course. The long-term remains bearish and we are still of the view that a move below the September low is the most likely course before the bear has run its course. The bonds may bounce over the very short-term but the medium-term picture has turned quite negative and an acceleration of the decline is not out of the realm of possibilities. The XAU may rally a bit more over the near-term but the decline from the May high does not look complete and a move back below the late November low is expected. 

Elliott Wave and Fibonacci

We have been and continue to approach the rally from September 21 on the S&P as a “b” wave related to the May-September decline. There are a couple of reasons for this count. The September 2000-March 2001 decline best counts as a simple a-b-c. The rally from March 22 to May 22 retraced about 50% of the preceding decline but a lot more than .618 of the decline from January to March suggesting that this rally was in fact related to the entire September 2000-March 2001. The decline from May to September was (is) a clean three-wave pattern in its own right so it cannot be counted as a “c” wave from September of 2000. As an aside, the January-March decline was almost perfectly equal in length with the September 2000-December 2000 pattern. As such waves “a” and ‘c” were equal and that is a very normal relationship. Moreover, the “b” wave (December 21-February 2) stopped just shy of retracing 50% of wave “a”, another strong relationship. We can count it the May-September decline as a second three of a double three from the September 2000 high. However, , the May 2001 decline did not have any sort of Fibonacci relationship with the decline from September 2000-March 2001. And given the relationships leading up to the May top we would expect that to be the case. Thus we are still focusing on the preferred count labeling the rally from September 21 as a “b” wave from May. The rally has thus far taken the S&P right up into the area of a .618 retracement of the decline from May and important resistance. However, “b” waves can and often times do retrace a lot more than .618 of the preceding “a” waves, In fact there are times, if the pattern is going to unfold as a flat, where the “b” wave can retrace nearly all of the preceding “a” wave. So a break of this resistance zone in and of itself would not invalidate the preferred count. The weekly chart from September has subdivided with the latest sub division occurring at the November 29 low, which did move below the low of the preceding week. On the weekly chart we can count the S&P as being in a fifth wave from September beginning on November 29. However, the daily and hourly charts within what can be counted as waves 1 and 3 are clearly corrective in both cases. Thus we are approaching the rally as a corrective one at this time. The only way we could count the November 1 advance as wave 3 would be if wave 4, ending on November 29 was an irregular. However, the only way to count the rally from September 21 as wave 1 would be if wave 2 was an irregular. Since waves 2 and 4 in a five-wave sequence have to alternate, and since in this case they don’t, this count is moot. The only way we can continue to support a bullish case is if we are in the early stages of a third of a third. If true than we will know soon enough as we should begin to see a strong acceleration to the upside. If we move much above the 1220 level on the S&P this could become a very real possibility but given the technical backdrop it is highly unlikely. Meanwhile, the rally from November 29 is so fare a thee-wave pattern on the daily chart. We are approaching this rally as either a “c” wave of a “c” wave diagonal triangle from November 1 or as the “a” wave of a third three of a triple three from September 21. Again, the 1214-1220 level come into play as that is where the rally from November 29 would be longer than the rally from November 1 if that rally is wave 3 and this is also the area where the rally from November 1 would be equal to the rally from September 21 to mid October. We have been approaching the post September 21 rally in the DJIA with the same idea as the S&P in that it is a “b” wave related to the May-September decline. The DJIA has slightly surpassed the .618 retracement of that decline but that alone would not detract from the count. However, the pattern from the January 2000 top is not as clean as the S&P from its September 2000 orthodox top. Also, the decline from May to September did have a Fibonacci relationship with the January 2000-March 2001 decline as it was close to 1.383 times the length of the first decline. Thus in the DJIA we can make the case that the rally from September 21 is in fact a larger degree “b” or X wave correcting the entire post January 2000 decline. The pattern from the September 21 low is nearly identical as the S&P both on the daily and weekly charts so for now we will approach the DJIA in the same way as the S&P.  On a short-term basis the rally from December 3 to December 5 was confirmed as a completed wave on the daily chart and as a five on the hourly chart of both the DJIA and the S&P. On the S&P we did have a very minor failure for wave 5 by exactly .25 points. Since the December 3 low never did take out the November 29 low it is possible that the rally from December 3 is a third wave on both the DJIA and the S&P from November 29. The low on Friday stopped in the area of an acceptable retracement for fourth waves and in fact there is a bit more room before that level is violated but not a whole lot. So it is possible that the post November 29 rally could unfold as a five. The NASDAQ 100 monthly chart can be counted as a big five-wave decline from March 2000 to September 2001. However, both the daily and weekly charts from the May 2001 peak to September’s low cannot by any stretch of the imagination be counted as impulsive. Thus we have to view this decline as a corrective one and not a fifth wave. It is possible that it is a “b” wave of an irregular from April and the current rally a “c” wave, however, it is very difficult to count the post September 21 as impulsive at this point so it is difficult to count it as a “c” wave. The NDX has moved just slightly past a .618 retracment of the May-September decline so it is possible that the rally is a “b” wave related to the May-September decline. There is also a possibility, albeit a slight one, that the decline into September was (is) a “b” wave of a triangle from April and the current rally a “c” wave of that triangle. This could ultimately allow for a big five-wave decline from March of 2000 but for now this count has to be viewed as only having the smallest of possibilities. The rally from December 3 was confirmed as a completed wave on the daily chart Friday. This rally can be counted as a five on the hourly chart. However, unlike the DJIA and the S&P it is impossible to count this rally as a third wave from November 29 as the November 29-November 30 rally was a clean three while it can be counted as a clean five on the DJIA and the S&P. So we either have a three-wave corrective pattern complete from November 29 or December 3 ended some sort of a triangle gong as far back as November 14. This latter count would also argue that the peak last week completed a more important top. The NDX weekly chart did not sub divide on November 23 so the pattern from September 21 is a three on this average. As such it is possible to count the rally from October 30 as a second three that is complete from September 21 but again impossible to count it as impulsive. Support: S&P: 1150-1151, 1140-1142, DJIA; 9985-9995, 9880-9893, NDX; 1648-1650, 1622-1629. Resistance: S&P; 1165-1167, 1175-1180, 1214-1220, DJIA: 10,105-10,115, 10,275, 10350-10,400, NDX; 1700-1706, 1738, 1800-1825. Trend changes for the next two weeks are as follows: December 12-13*, December 20-21**. * denotes important.       

Bonds

The bond market had a weak rally into the early part of last week. It did not come close to important resistance but it was enough to confirm the decline from November 1 to November 27 as a completed wave on weekly chart and as a five on the daily chart. The decline last week took the bonds substantially below the November 27 low and it is likely that wave 3 or “c” from November 1 is underway. There is an outside chance that the decline last week is a “b” wave of an irregular second wave from November 27 but this is a very low probability. And if it is we need to hold very close to current levels. At this point we are operating with the idea that the November 1 peak was either a deep second wave from October 1998 or the post January 2000 advance was the last wave completing a large corrective pattern from 1980. In either case the bonds would be in the early stages of a fairly important decline. The fact that we are so far declining in what looks to be very impulsive fashion supports either count and we have to be aware that we could be in the early stages of a major decline. The drop from November 1 has wiped out in 5-weeks just over half of the gains from January 2000 to November 2002.In addition, the bonds have broken a trend line drawn off the January 2000 and May 2001 lows and did so leaving little doubt. The daily range oscillators are modestly oversold. The daily trend oscillators are negative. Short-term momentum is down but could support a modest bounce. The weekly range oscillators are negative but not yet oversold. The weekly trend oscillators are negative and medium-term momentum is clearly negative. Sentiment indicators are improving. The Rydex ratio of is on the plus side of the ledger. In addition, the commitment of traders report is showing a slight net long position from the commercial hedgers. However, the small trader is also net long and that is a negative. Consensus Inc. has improved to 39% bulls but that is only neutral, while market Vane is still at a bearish level at 58% bulls. Support: 97-97 20/32, 94 24/32-95 10/32. Resistance; 101 17/32 +/- 6/32, 102 29/32 +/- 11/32. Please note that all price projections are based on a constant bond chart and not any particular futures contract. The constant bond chart is currently running 2/32 either side of the March 2002 futures contract. The short-term is getting oversold again and we could get a bounce towards resistance at anytime. However, the medium-term momentum picture is clearly negative and the potential that we are in the early stages of a third wave is high and that could usher in an acceleration to the downside. We are going to remain neutral on the short-term while moving to negative on the medium-term. 

XAU

The XAU has continued to rally and decline in three-wave patterns on the daily, weekly and monthly charts going back to the October 2000 low. This leaves open the possibility that a large triangle of some sort is in development. This could if correct lead to a final washout low to complete the long-term bottoming process that began in October of 2000. Short-term momentum is slightly positive and could support as bit more rally. Medium-term momentum is still headed lower but showing signs of stabilizing and perhaps bottoming soon. Some of our sentiment indicators have shown some improvement. They are not bullish as yet but are clearly no longer bearish. Support: 50-51, 45-45.25, 42,42.24. Resistance; 55.50-56, 60-60.50. The short-term is still OK so a bit more rally to initial resistance is a possibility but this is not enough to move from neutral. The medium-term is improving but not enough and we remain negative. The long-term is neutral. 

Indicator review

November, as we had expected it would in the last issue, was a very strong month across the board. December has also started out in a very positive mode with the first week of the last month of the year showing across the board plus signs. The market has had a very strong tendency to rally early in a new month and from this perspective it did not disappoint. The market has also had a tendency to make some sort of a top early in a new month. As far as this is concerned the jury is still out. The rally last week has moved both the DJIA and the S&P into the area of some important resistance. Both averages are also close to their declining 200-day moving averages. A break above them would not be a surprise but in our view would also not be a bullish factor given the fact that they are in a down slopping pattern (fro more on this see the last report of November 26). The rally mid week was quite strong in terms of points gained. Volume flows did improve a bit and the very strong rally on Wednesday was accompanied by a huge spike in volume, which hit its highest one-day total since late September. As we pointed out in that nights report, expanding volume on rallies is usually a plus. However, often times a strong surge in volume after a trend has been in place for a while, and this one is over two months old, tends to mark a turning point. Breadth was also solid last week and has been throughout the rally. The new high list also expanded and finally confirmed price on Wednesday reversing weeks of poor action. The new lows have eased but not by a lot. The high/low indicators are neutral. The Russell 2000 has started off December in very strong fashion as it gained over 4% last week. The short-term is neutral but also very extended with a number of divergences beginning to develop, making it vulnerable to a reaction. We are going to remain neutral on the medium-term but with a still modestly positive bias. The Value-line also had a strong initial start in December adding nearly 4% as well. We are neutral on the short-term but as is the case with the Russell it is extended and beginning to show a number of potential divergences. We are neutral medium-term but with a still modestly positive bias. The NASDAQ Composite and the NASDAQ 100 gained nearly% last week. The short-term is neutral but showing a lot of cracks and is in a position to correct some. The medium-term is also neutral but with a still positive bias. The DJTA continued to push higher and closed the week with strong gains. We are going to remain neutral on the short-term as it is very extended. The medium-term is slightly positive. The DJUA and UTY ended the week higher. They are neutral short-term but close to going positive. The medium-term is negative but stabilizing. The long-term remains down.

Momentum    

The momentum picture has weakened further over the past two weeks. The breadth and volume oscillators are showing serious divergences going back to early October. The 3-day oscillator is in a similar position as is the McClellan oscillator as they have all failed for the second time to confirm new highs in price. The 10day Arms is neutral as is the open 10. The 5-day Arms is negative as it is overbought while the 21-day Arms is neutral. The new 10-day Arms is back to overbought and not far from issuing another sell signal. The daily range oscillators are near overbought and are also diverging. The daily trend oscillators are neutral but very, very weak. The weekly breadth measures are near overbought and also showing some weakness as they turned down last week in spite of the rally. The weekly range oscillators are neutral. The weekly trend oscillators are positive.

Sentiment 

The sentiment model was unchanged for the past three-weeks at a modestly negative +7. Investors Intelligence reported a drop in both bulls and bears. The latter is at very negative levels seen near important tops. Market Vane eased slightly last week as did Consensus Inc. The 4-week moving average for Market Vane is near bearish and at its most negative level in nearly two years. The 4-week moving average for the Consensus Inc. survey is bearish and near levels seen at nearly every important top of the past 2 plus years. The American Association of Individual Investors (AAII) showed a drop in bulls last week. However, that drop came from the second highest reading in its 14- year history and the bull/bear ratio is still quite negative as is this measure of public sentiment. The insider sell buy ratio has improved and basis the 8-week moving average is near bullish. NYSE members were on the sell side for the latest reporting period reversing 3-weeks of buying. This indicator is slightly negative. The CBOE put to call ratio basis the 10-day moving average has improved to neutral and given the rally this is a plus. The OEX ratio is bullish. The Rydex ratio is near the lower end of negative. The VIX and VXN did move up a bit late last week but from quite low and negative levels. The VIX was below 24 and not far from its May and August troughs. The VXN is still below its May level. They both could move lower but they are already negative and flashing big warning signs.

Summary and Conclusion 

Back in early October a number of breadth and volume related indicators reached a strong enough overbought reading to be viewed as a partial thrust signal. We had stated back then that the thrust was not strong enough to be considered a kick off to a new bull market such as we saw coming off the October 1998 low. For example, in October 1998 the McClellan oscillator moved well in excess of +250 only 8 days off the price low. In October it reached a modest overbought reading of +168 but 15 days after the price low. This did not even exceed the +187 reading seen in late December of last year. We did go on to say that although we did not see this as a kick off to a new bull market, these indicators did reach levels that most likely had positive medium-term implications and could be the beginning of a rally lasting for 2 months or so. This was not unlike the rally we saw in January following the December 28 thrust reading. Last week the rally from the October thrust was near two months old and the rally from September 21 about 2 ½ months old. What we have seen from these indicators is classic as they have, since their October peak, made two lower highs while price has made two higher highs, setting off what we cal triple divergences although technically there are only two divergences. This is very a typical of patterns following medium-term thrust signals and the time frame is also correct. We saw a similar pattern from the volume and breadth oscillator leading in the February-April 1998 time frame. That led to a peak before the final peak but also set the stage for an important top. In some cases a triple divergence occurs near important medium-term tops and in some cases what we get is what we saw in 1998. A top that lasts for a while followed by one last final rally. This final rally produces a modest overbought reading with these indicators actually moving above the third peak and occasionally the second peak but never above the thrust signal peak. What we have now in place is the set up for a good-sized decline. Whether this is the top of the post September 21 advance or not is hard to say. Given the overall technical backdrop we can make a case for a top now but can just as easily make a case for the pattern to play out like 1998 but without a new all time high only a new post September 21 high. Some sentiment indicators such as the Investors Intelligence survey and the American Association of Individual Investors are at very weak and negative levels. The percentage of bears from Investors Intelligence has dropped to levels seen at very important tops over the past several years. And just last week the AAII percentage of bulls was the second highest since the survey began in 1987. The only higher reading was seen in January 2000 just as the DJIA was making its peak. The overall picture from these indicators can of course get a lot worse but they already are at negative levels. The Rydex ratio is on the negative side of the equation but not excessively so while the CBOE put to call ratio basis the 10-day moving average is only neutral. One thing supporting the bullish case was the fact that late last week on only some marginal weakness and following two explosive up sessions the CBOE put to call ratio moved up to near bullish one-day readings. Also following Wednesday’s strong gains the asset level in Nova, a bullish play on the S&P did not move up at all. This is December and December is notorious for a lot of cross currents due to end of year strategies. And it is indeed possible that this could be affecting some of these indicators. However, on face value we have to view the action of these two as somewhat constructive and as an indication that the downside risk from here could be not as bad as it would given the very negative momentum backdrop. We do believe based on our indicators that the market is in the process of completing at least a short-term top. However, as bad as the indicators are, there is enough positives left to support the possibility that the top we see coming is not yet in place. As such at least for now we are going to remain neutral on the short-term but with a more cautious bias. Medium-term we see the rally as having more to go but also see it as being in the latter stages. A lot of key indicators are deteriorating quickly and we see the rally coming to a head soon. It is too early to go short and too late to go long. We will remain neutral. Long-term we remain bearish. For more on this see the November 26 report. 

SPY, QQQ and Futures traders are flat Stand aside for today. Rydex switchers are holding a 5% Ursa position. Sell only if the S&P is above 1192 at anytime on Monday. 

The bonds have broken below the rising trend line drawn off the January 2000 and May 2001 low. A rally back towards that line is possible over the near-term but we also see the possibility of the decline entering an acceleration phase. 

 The chart above shows the breadth oscillator from early 1998 to the July 1998 top. Note the medium-term thrust signal in February followed by two lower peaks into the early May high. The S&P peaked in April. We corrected for five weeks and that was followed by rally into July. That latter rally produced an overbought reading taking out the peak of the two divergent peaks but not the thrust peak. Further on the left of the chart you can see a similar pattern leading into the July 1997 top.

The chart below is the current breadth oscillator. Note the similar pattern in place now.

The McClellan oscillator on the NASDAQ is showing a similar divergence to the pattern leading into the May top. However, there is one notable difference as the peak on of the post September 21 rally did not come close to where it did following the rally off the April low. In addition, note the bullish divergences leading into the April low and also at the December 2000 low. Note how at the September low there are no divergences.

The 4-week moving average of the consensus Inc Survey is at levels seen at the last 3 important tops since January 2000.