BI-WEEKLY MARKET RECAP and TECHNICAL PERSPECTIVE

By: Larry Katz

April 23, 2001

Capsule

Sentiment measures are still OK but the overall technical picture has gone from positive to negative, and by some measures extremely negative. The technical barometer has just recorded its most negative reading ever. Short-term the rally looks to be near complete and a decline is eminent. While the March April low(s) may have completed a medium-term low (we’ll know by the nature of the expected decline) long-term the bear market remains in force. The bonds may be getting close to a bounce but remain in a strong medium-term downtrend with lower prices in store. The XAU is in a bullish position but needs to rally strongly through resistance above 5.25 to keep the pattern positive.

Elliott Wave and Fibonacci

The best laid plans of mice and men sometimes get thrown astray and so it was with my preferred wave count as it related to the rally from March 22 in regards to the S&P. The move above the March 27 high eliminated any possibility that the rally from March 22 was a fourth wave from the January 31 peak, confirming that the decline from January 31 is a completed wave all into itself. The more important question to answer is whether the rally from March 22 is correcting only the post January 31 decline or possibly the entire post September 1 decline. Since I have been counting the January 31 decline as a “c” wave from September it would seem that the logical choice is that the S&P is in the process of correcting the entire decline from September. This is the most logical choice based on the preferred wave count. However, there is also a very distinct possibility that the January decline was a “c” wave but a “c” wave from the November 10 high not from the September 1 high. This does not invalidate the possibility that the entire post September decline is not complete as the wave structure under this count does allow for the completion of a double three as opposed to a flat as the preferred count would have it. However, the alternate count does allow for lower lows as a double three can turn into a triple three. This is of course assuming that the January decline is a five. I can count it as a five and so we have to go with that count, however, it can just as easily be counted as a three and that further complicates the pattern as that decline could be counted as a third three of a triple three from September or only as an “a “ wave of a larger pattern. There are obviously a lot of possible ways to count the pattern not only from September but also from the January peak. This has turned out to be the rule not the exception when dealing in wave counts, which has led me to relay a lot on Fibonacci price relationships to help in determining the correct count. The 1107 level was an important point on the S&P as that was the level in which the January 31 decline was equal to the decline from September 1 to December 21. The fact that the S&P moved well enough below that level was enough to expect that either the post January decline was either extending or that the preferred count was not correct and to begin to look elsewhere. Frankly it is still possible that the post January decline is not over. We are dealing with a three-wave rally from March 22 on both the daily and weekly charts, which is still below the .618 retracement of the post January decline and this leaves open the possibility that the rally is either a second wave or “b” wave from January that is nearly complete. However, Fibonacci relationships from the January peak with both the September to October and November to December declines argue that the post January decline, if not a “c” wave from September is indeed a “c” wave from November. As it stands, the decline from November 6 to March 22 was 1.618 the September 1 to October 18 decline at 1075. In addition, the decline from January 31 to March 22 was 1.618 the November 6 to December 21 decline at 1085. The actual low was 1081. Thus the second three was almost exactly 1.618 the length of the first three and wave “c” of the second three was almost a perfect 1.618 the length of wave “a” of the second three. As mentioned above, it is still possible to count the January-March decline as the first wave of a larger pattern and given the pattern from March 22 the rally can be counted as either a second or “b” wave from January. However, the post March 22 rally can just as easily be counted as the first wave of a larger corrective pattern with March 22 completing the “a” wave of intermediate wave 4. Short-term the daily chart shows the April 12 low to be a whole new wave from April 4. The April 12 rally can be counted as a completed five-wave advance on the hourly chart on Friday but has not yet been confirmed as complete on the daily chart. A move below Friday’s low of 1234 would do the trick. This rally is either an extended fifth wave from April 4, a “c” wave from April 4 or a third of a third from April 4. At this point I have no high confidence count either way but if it is a third of a third we need a couple of 4’s and 5’s to complete the pattern. In any case, the rally from March 22 to March 27 was a confirmed three-wave pattern so the rally from April 4 is either a “c” wave or a second three from that low and as such the post March 22 rally is not the beginning of a new impulse move but part of a corrective pattern. The pattern from January 2000 on the DJIA is clearly corrective and is best counted as a double three with the first three ending in October and the second three a simple flat coming out of an X wave triangle ending on March 8. The alternate count  is that February 6 completed a complex  “b” wave from October 2000 and that the decline into March 22 was an “a” wave of a larger pattern. The DJIA has, however, retraced a good deal more than a .618 relationship with the February decline making it more likely that the primary count is correct. However, this does not invalidate the alternate count as the “b” wave of a flat could very well retrace a lot more than .618 of the preceding “a” wave.  If the alternate count is correct we could be very close to a large “c” wave that will carry the DJIA below the March 22 low. In any case, I am counting the rally from March 22 as a “b” wave either from January 2000 or February-March 2001. Friday the DJIA moved below Thursday’s low and that confirmed the rally from April 12 as a completed wave on the daily chart. That rally can be counted as a five on the hourly chart but not easily so. However, I will give it the benefit of the doubt and count it as a five. As such the rally is either an extended fifth wave from April 4, a third of a third from April 4 or a “c” wave from April 4. Under the first two counts the rally from April 4 would be counted as a “c” wave from March 22. If the latter than the post April 4 rally is best counted as a second three from March 22, with the first three ending on April 2. A move above Friday’s high of 10,702 would eliminate both the “c” wave and extended fifth wave count and argue that the post April 12 advance was 3 of 3 from April 4.     There is a cluster of Fibonacci resistance not too far from last weeks high. First off, the .618 retracement of the decline from January 2000 is close to 11,740. In addition, the rally from April 12 would be equal to the April 4-April 11 at 10,740 and lastly the rally from April 4 would be 1.618 the March 22-April 2 rally at 10,809. I had been counting the decline from January on the NDX as a fifth wave from September. This is still a valid wave count although it did exceed by a side enough margin important Fibonacci relationships with wave 1 (September 1- October 18) of the pattern. The NDX confirmed the decline from January as a completed wave on the weekly chart. However, the monthly chart remains down and until that turns up it will be possible to see the post September 1 decline extend and that leaves open the possibility that the January decline was not wave 5 but only 3 of 3 from September. Again the key is the monthly chart. A move above 2029 would take out the high of March and confirm the pattern from September as a completed wave on the monthly chart. Until that occurs we can not rule out the possibility that the rally from April 4, as dynamic as it has been, is nothing more than a fourth wave. I am leaning in favor of the bullish count but we still need to get conformation from the price structure. The rally from April 17 can be counted as having completed five waves up on the hourly chart at Friday’s high. However, the daily chart is still positive and until that confirms the rally may extend. A move below Friday’s low of 1901 would confirm. The rally from April 17 is either a third of a third, an extended fifth wave or a “c” wave from April 4. There is strong Fibonacci resistance in the 1960-1980 zone, which did stop the rally on Friday. This is where the rally from April 17 was equal to the rally from April 4 to April 11. It is also where wave 5 from April 17 was 1.618 wave 1 from April 17. In addition, the rally from April 17 was 1.383 the rally from April 6 to April 11 in this zone. The April 6 to April 11 rally may be counted as wave 1 of 3 if the pattern is subdividing. The bottom line is that we have a completed five-wave pattern from April 17 at important resistance and a decline of at least short-term is likely underway. Support: S&P; 1232-1233, 1216-1218, 1192-1195, DJIA; 10,400-10,415, 10,225-10,240, NDX; 1901, 1836-1840, 1747-1753. Resistance: S&P 1258-1261, 1268-1272, DJIA; 10,730-10,740, 10,810-10,835, NDX; 1960-1980, 2029. Trend changes for the next two weeks are as follows; April 22, April 25*, April 29, May 1-2*, and May 6*. *denotes potentially important dates.

Bonds

The post diagonal triangle thrust has continued on the bonds and the decline is taking on clearly impulsive characteristics. It is possible in fact to count the decline from March 23 as a nearly complete five wave pattern on the daily chart but given the price performance it is more likely that the bonds are in the latter stages of wave 3. While we are a long way from confirming that the bonds have  entered wave 3 or “c” from the October 1998 there is enough evidence to suggest that this is indeed the case as the depth of the post March decline has confirmed that the rally from January 2000 is corrective. The bonds have also broken a major trend line drawn off of the January, May and November 2000 lows. The daily range oscillators are oversold but have also confirmed the price lows and that is consistent with third waves. The daily trend oscillators are negative and still accelerating and short-term momentum remains negative. The weekly range oscillators are not yet close to oversold and remain very negative. The weekly trend oscillators are negative and just beginning to break down and accelerate. Weekly momentum is negative. Sentiment has improved with Market Vane at  47% bulls and Consensus Inc. at 30% bulls. This is still only neutral, however, especially given the extent of the decline and this needs to improve a whole lot more before it will be a positive factor. Support: 98 25/32, 97 30/32-98 06/32 and 95 26/32 +/- 24/32. Resistance: 100 16/32-100 24/32, 102 10/32+/- 7/32. Please note that all price projections are based on a constant bond chart and not any particular contract. Currently the constant bond chart is running 6 32 below the nearby futures (June).  Short-term the bonds are getting very oversold, which could set up for a rally but clearly the pattern from March 23 is not over and lower prices even short-term are expected. The move to neutral last week on the short-term was premature and although we could bounce I am moving back to bearish. Medium-term I remain negative and long-term very bearish.

XAU

The rally from April 3 has carried the XAU to slightly above the .618 retracement of the decline from February 27. The rally so far is not impulsive but could develop as impulsive provided that the XAU continues to rally. Thus there is very little room for weakness from current levels and a strong move above last weeks high of 53.37 should occur soon if indeed the pattern is going to unfold as impulsive. Meanwhile the fact that the decline from February 27 is showing a corrective pattern on both the daily and weekly charts adds some support to the idea that a new up wave is getting underway or at least the decline is not impulsive. Short-term momentum has turned up while both weekly and monthly momentum remain positive. However, the position of the weekly and monthly indicators are borderline and higher prices now are needed to keep them rising. Sentiment on gold futures low and from a contrary perspective that is bullish. Support; 51.50, 48-48.50. Resistance: 53.50, 55-55.50. Sentiment and momentum is positive with the XAU in its best position to rally since the early February low. I moved to bullish late last week across the board and remain so. However, in its important that we continue to rally or the indicators could reverse.

Indicator review

The first week of April started off poorly but the past two weeks the averages have come roaring back led by a resurgence in technology stocks. The DJIA has recouped a good portion of its post February losses and has moved up into some stiff resistance. The S&P after lagging for most of the year has done well and on a percentage basis from the April 4 low has kept pace with the DJIA but the past two weeks and especially late last week has begun to show a bit of positive relative strength. Wednesday’s strong price surge was accompanied by a very large expansion in volume reaching the second highest level in NYSE history. Volume on Thursday and Friday was still good but did slow but price/volume relationships have moved from negative to neutral. However, it will be important to see volume continue to expand on rallies or the pattern could turn negative quickly. As it stands Wednesday had the look of climactic action in reverse and while volume has been good the up to down volume ratios have been nothing exciting. Breadth was positive but clearly did not keep pace with the averages lagging badly especially late in the week as it went dead. The weekly statistics were OK but that was about it and clearly not as strong as the averages. The high/low statistics weakened late in the week as the number of new highs, which on Wednesday confirmed for the post March rally dropped in half on Thursday and contracted further on Friday. The high/low indicators are still OK but the easy comparisons are going to disappear over the next few days. The strength in technology and the NASDAQ in particular looks to be more than just a one to two day wonder and we could be in the early stages of a shift in relative strength it is long overdue to say the least.  The Russell 2000 has rallied smartly the past two weeks but is lagging and into some important short-term resistance and is slightly overbought. I am neutral short and medium-term. The Value-line has done quite well recouping about 2/3rds of its loss from January. The short-term is neutral and getting overbought. The medium-term is neutral. The NASDAQ Composite and NASDAQ 1000 have been the clear leaders from the April 4 low rallying 36 and 48 percent respectively. However, in spite of the impressive gains neither average has moved above their March high and neither have been able to retrace even half of what they lost from the January peak. That sure does put into perspective how much carnage was done in the technology sector. As pointed out in a previous report both of these averages could double from their April low and not recoup half of what they lost from their March 2000 peak. The short-term is neutral overbought and close to going negative. I am going to stay neutral on the medium-term but could turn bullish on a correction. The DJTA has had a good couple of weeks participating along with the rest of the averages. The short-term is neutral but looks higher. The medium-term is neutral.  The DJTA and UTY were able to break out above important resistance but that break out was short-lived as both moved back below that resistance late last week giving the impression of a false move. I am going to remain neutral on both the short and medium-term. Long-term I remain bearish.

Momentum

The breadth oscillator reached mediocre overbought levels and turned down failing so far to come close to thrust levels. The volume oscillator moved to extreme levels of overbought and at Wednesday’s peak hit its most overbought level in the 31 years of data that I have. This is as extreme as it was on March 22 when it hit its most extreme oversold level. The 3-day oscillator turned lower after reaching overbought levels on Wednesday and is on a short-term sell signal. The McClellan oscillator hit modest overbought readings on Wednesday as it poked above the +100 level. However, it failed to generate a real positive momentum reading such as it did coming off the 1998 low when it moved to over +250 only 12 days after the price low. The 10-day and open 10-day Arms are now overbought or close and are negative. The 5-day Arms is overbought and the 21-day Arms is neutral. The new 10-day Arms has issued its third sell signal in a little over a week at Friday’s close. The daily range oscillators are slightly overbought. The daily trend oscillators are positive. The weekly breadth oscillators are neutral. The weekly range oscillators are neutral but close to oversold. The weekly trend oscillators are neutral but close to going positive. The technical barometer has turned extremely negative (there will be more on this in the summary and conclusion section).

Sentiment

The sentiment composite at +10 is neutral but has slipped the past couple of weeks. Investors Intelligence has shown some good improvement with the one-week bull/bear ratio about dead even. However, it never has come close to levels seen at important market bottoms and would still have to be rated as only neutral. Market Vane and Consensus Inc. moved up a bit last week but both are still positive. The American Association of Individual Investors (AAII), which was improving a couple of weeks ago has already weakened as it reported a sharp rise in bulls and drop in bears. The indicator is at best neutral.  The commitment of traders report on the S&P futures has improved the past couple of weeks but is still showing a huge net short position on the part of commercial hedgers and a fairly large net long position on the part of the small speculator. The commercial hedgers are the equivalent of the members on the NYSE or corporate insiders and the small speculator the usually wrong public. This indicator remains bearish. The NYSE members on the other hand continue to be net buyers and this indicator is bullish. Corporate insiders have eased a bit their level of selling but the indicator remains bearish. The Rydex ratios are not fully negative but have weakened a lot. Moreover, the levels of assets in the Arktos and Ursa funds (bearish plays on the NDX and S&P respectively) have dropped sharply and are a stone’s throw from where they were at the late January high. The CBOE put to call ratio basis the 10-dfay moving average has eased considerably the past week and although not close to excessive it is at its lowest level since March 8. The OEX ratio has moved back to negative.

Summary and Conclusion

The rally the past couple of weeks has been quite impressive and in the case of the NASDAQ, long overdue. As that old expression goes “You can only stretch a rubber band so far before it snaps”, and the further you do stretch it the further it will snap and on April 4 it got stretched pretty hard. The intensity of the advance has produced a couple of very interesting developments. For one thing it has produced a chorus line of calls that the bottom is in. Now instead of hearing and seeing reports on how to finds a bottom we can hear reports on what to buy for the new bull market. I said it once and will say it again, bear markets do not end when everyone is looking for them to end, no they end when no one wants to own stocks anymore. Furthermore new bull markets especially early on are not greeted by reports on what to buy but on reports of what to sell. In other words, they are not believed. Some of the sentiment surveys such as Market Vane and Consensus Inc are still in fairly good shape and the last couple of weeks has actually seen some improvement in the Investors Intelligence survey. However, the latter never came close to levels seen at important market bottoms such as 1998, 1994 or 1990 let alone 1982 and is no better than neutral. Short-term sentiment as measured by the 10-day moving average of the CBOE put to call ratio is neutral but dropping rapidly. Moreover, the Rydex ratio has moved up from near bullish to low neutral.  However, the asset levels in the bear funds have dropped like a rock and are not too far from where they were in late January-early February, which were record lows. Momentum meanwhile has done a complete about face moving from deeply oversold to overbought in the course of 4 weeks. Volume related indicators have in fact reached thrust levels with the volume oscillator at extreme record overbought levels. This could be construed as a medium-term kick off expect for the fact that the breadth related measures are not even close. They did reach overbought levels late last week but only modest ones and that was nowhere close to levels that could be even remotely viewed as a thrust. For example, the McClellan oscillator hit a modest +102 on Wednesday. If we measure from the price low of the S&P on a closing basis it was 11 trading days. In 1998 the McClellan oscillator moved from oversold on October 8 to +257 by October 20. That was only 8 days off the low and that is what a momentum thrust is all about. The Arms indexes have moved from extreme levels of oversold to overbought and the new 10 Arms has, as of Friday’s close issued its third sell signal in a little over two weeks. The technical barometer, which is reported in this report in the indicator review section has reached an important level. The barometer is a combination of 10 indicators from momentum to sentiment and one monetary measure. There are two calculations for the barometer. The inside position is calculated every Friday and the indicators are either bullish, bearish or neutral. A reading of –4 is a bearish warning and +4 is bullish. A –5 or greater is negative. In the outside position the indicator is either bullish or bearish, there is no neutral reading. It is not based on Friday’s close but at any point in which the indicator moves from bullish to bearish or visa versa. The barometer was designed to forecast price changes from one to three months out it is not an exact timing indicator but a leading indicator. I have kept it since early 1994 and it has had a very good record of calling important turns in the market. In 1994 for example we had a couple of periods where the outside score reached a very bullish +6 and in 1998 it hit a +4 on a couple of occasions. The best we have seen in the past 13v months was a moderate +2 in September of last year. Even during the carnage of the first quarter we saw no better than a 0. The inside score did reach a +2 reading on March 23 but that is a far cry from a +4 or +5 seen at important lows. At last weeks calculation we have reached the most negative reading I have seen from this key market indicator. The inside position is a –6 and the outside position a –10. That is the worst it can be as all 10 indicators are negative.  Prior to this reading the next worse signal came in February of 1998 when we had two weeks of a –8 in the outside position accompanied by a –6 reading on the inside position. This of course occurred several months after the October 1997 low, which by the way never produced much of a buy signal from the barometer. However, it did take 4 months to issue a sell signal while this time at=round we got it in a little under a month. The market did rally further and made its top internally at least in April and the S&P in July. This may give us some time on a medium-term basis but the message is clear that the long-term picture remains bearish and the signal we are receiving now is extreme. There is always an outside chance that the market may just go sideways to correct the recent overbought condition. We have seen that plenty of times especially if the market is in the beginning of a medium-term advance. However I see enough from both the momentum indicators as well as the short-term sentiment measures to move from neutral to bearish on the short-term. The nature of the decline I am expecting will be important to the medium-term. If it is too deep or develops in an impulsive fashion it could turn the medium-term negative. If it is orderly it could very well be a part of a medium-term advance. For now I will stay neutral.  In any case the rally should be viewed as a bear market rally only and not the beginning of a new bull market or a continuation of the old bull market. I remain bearish on the long-term.

QQQ traders sold short a ½ position on Friday at 49.25 per the hotline. They closed at 48.40. Keep your stop at 50.82. Lower the stop to 49.54 on any move by the NDX below 1900. Rydex switchers are holding a 20% Precious metals and 10% OTC position. Make sure to call the Noon Pacific hotline for any changes. The morning hotline will be on at 7:15 AM Pacific time.       

The chart below shows the McClellan oscillator and the S&P. Note the huge spike from the October 1998 low taking the oscillator from a –140 reading to an extreme overbought +253 in 8 trading days. The rally from April 4 has lasted 11 days and the best we got on the oscillator was a +102, which is a very weak reading.

The chart above on the S&P shows the two wave counts discussed in the Elliott wave section. In addition to important Fibonacci resistance just above current prices there is also resistance from a couple of trend lines. Trend line a defines the neckline of a large head and shoulders top and trend line b connects the top of September 1 with the January 31 top.

The daily chart of the NDX shows the 3 possible wave counts from the April 4 low.

The chart above shows the assets in both Arktos and Ursa. Note how quickly they have moved lower and are in very close proximity to where they were in late January. Note also how far they were from where they were in late April-early May of last year prior to a multi month rally.