BI-WEEKLY MARKET RECAP and TECHNICAL PERSPECTIVE

By: Larry Katz

August 20, 2001

Capsule

Short-term indicators are improving and the market may be setting up for a rally. However, any rally from current levels and current technical conditions should be viewed as only a short-term phenomena and not the start of any sustainable advance. Long-term the bear remains firmly in control and there is no bottom in sight. The bonds are stretched short-term but do have some momentum behind them and some further gains are likely. The XAU is close to important resistance and sentiment has turned negative. A move back towards the early July lows is expected soon.

Elliott wave and Fibonacci

Our long-term count on the S&P has not changed. A long five-wave advance from the 1987 high completed either at the March or September 2000 peak and rather close to our long-standing price target in the 1550 area. We are now favoring the idea that September was in fact the orthodox peak from 1987, not March. The decline from September to the March 22 low was a clean three-wave pattern on the daily, weekly and more importantly the monthly chart. The pattern is best counted as a double three but it is possible to count the decline from late January to March 22 as a five on the daily and weekly chart. It is not a clean five but it is possible and that leaves open the possibility that the post February 2 decline was a “c” wave from September and not a second three In either case s our  preferred count is that it s the low in March represented the “a” wave of  intermediate wave 4 from 1982. This was reinforced by the fact that the March-May rally is best counted as a corrective three-wave rally and not the beginning of a new impulse wave. Since we are in a fourth wave it is possible that the pattern could ultimately evolve as a triangle and under that scenario it is possible the price low seen in March was all of wave “a” and will not be broken. However, that is not our preferred count nor for that matter does the long-term technical backdrop support such an outcome but nonetheless it is a possibility. The rally into the May 22 peak moved well above a .618 retracement of the decline from January to March but did stop right near a 50% retracement of the entire post September 2000 decline. And that supports the idea that the rally into May was directly related to the entire post September decline. In the last report we were approaching the decline from May 22 to July 24 on the S&P as a double three. On both the daily and weekly charts that decline was and still is best counted as a seven wave corrective pattern. Last week the S&P moved below the July 24 low. The decline from August 2 is so far a three-wave pattern on the daily chart. It is possible but highly unlikely that this decline is a “b” wave of an irregular from July 24. Therefore we need to approach the post August 2 decline as a new wave down from the May 22 high. As such it is either a “c” wave from august 2 or all or part of a second three. While the May 22-July 24 decline is definitely a corrective pattern, it is possible to count the decline from May 22 as a five going into the July 6 or 11 low and that the pattern from either of those lows into the August 2 peak was a second or “b” wave completing on August 2 (this count is shown on both the daily and weekly charts on the web site). The fact that this can also be confirmed on the weekly chart adds some validity to the count. However, if this is the case then it is imperative that the post August 2 decline unfold as impulsive. Even if the latter count proves correct, we are still faced with what is still a three-wave pattern and that leaves open a couple of possibilities as to where the post May 22 decline fits within the pattern from September 2000. the most obvious is that it is a “b” wave from March and once complete will be followed by a “c” wave rally. That could allow for a rally back above the May 22 peak before wave “c” from September 2000 really unleashes. However, that  “c” wave could be part of a larger “B” wave triangle from March and in that case a move back above the May 22 high is not going to occur. The other possibility is that the May 22 high was it as far a rally and that the decline from May is either the “a ‘ wave of a second three from September 2000 or part of a larger “c” wave from September 2000. If we are to get a five from August 2 we have a some room to rally over the near-term but not a lot as we are still faced with only three-waves down from August 2. The hourly chart from August 14, which would be counted as wave 2 from August 2 does not yet count well as a five and too suggests that any rally should be very minor and short lived if indeed the post August 2 decline is to develop as a five. We are still dealing with what is still a three-wave decline on the DJIA from May 22 on the weekly chart. As we have discussed in the past, it is possible that the pattern from June 26 was some sort of a fourth wave triangle from May 22 and a move to new lows below the July 11 low would give the weekly chart a possible five-wave pattern to the downside. While the DJIA has not broken below that low it did last week marginally below the July 24 and August 10 low. It is possible to count the mid weeks high last week as the “e” wave of this triangle but the pattern is not the cleanest one we have seen.  The hourly chart from Tuesday is not impulsive. It is possible that a five could evolve and that will be necessary if indeed it is a fifth wave from May. The other possibility for this particular count is that the triangle is still in force and we have just completed a very complex “b’ wave of that triangle. Of course the other possibility is that the entire post May 22 decline is not going to unfold as impulsive but corrective and then of course the triangle discussion is moot. On a longer-term basis we are approaching the DJIA in the same way as the S&P. We had thought back in April that there was a possibility that the March low in the DJIA was a fourth wave from 1987 and that would have allowed for a move to new highs to complete the pattern from 1987. However, the rally from March to May was confirmed as a three on the weekly chart and that eliminated the possibility that it was part of a fifth wave from 1987. The decline from January 2000 into the March 2001 low is best counted as the “a” wave of intermediate wave 4 from 1982. The rally into May was either a “b” or X wave from January 2000. As is the case with the S&P, it is possible since we are in a fourth wave that the entire pattern unfold as a triangle and if that is the case it is possible that the May high was wave “b” and that we are currently in wave “c”. While this is a possibility it is not our preferred count. It is also possible that the pattern from March is going to itself unfold as a larger “b” wave triangle with the May high wave “a and the current decline wave “b”. A lot is going to depend on whether we get a five-wave decline from May so the next week or so will be important. The NDX monthly chart from March 2000 shows a clean three-wave pattern. The rally into May from April’s low did not even come close to a .383 retracement of the decline from September of 2000. That leaves open the possibility that the rally into May was all or part of a fourth wave from March 2000. Even if the NDX were to move back above the May high it would still be quite possible that the entire rally from April was a large degree fourth wave from March 2000. we have been approaching the decline from May as corrective and it still best counts as such. However, it is also possible to count the decline as a series of 1’s and 2’s and that from August 2 we are entering a third of a third (this is shown on the web site on the weekly chart). The post August 2 decline is so far a three-wave pattern on both the daily and hourly charts with both sub waves counting as fives on the hourly chart. There is some room to rally over the near-term but if we are indeed in a third of a third from May 22 the rally if we do get one should be short and sweet and the possibility of an acceleration from here is high. An alternate possibility similar to the DJIA is that the decline from May 22 in the NDX is a “b” wave of a larger fourth wave triangle. It seems guite likely that this coming week should be important to a number of wave counts. Support S&P; 1135-1141, 1078-1083, DJIA; 10,100-10,120, 9950-9975, NDX;1470-1480, 1445-1452, 1363-1375. Resistance: S&P; 1170-1172, 1182-1185, DJIA; 10,313-10,321, 10,490-10,505, NDX; 1570-1574, 1611-1618. Trend changes for the next two weeks are as follows: August 22-23*, August 27  August 31-September 4**.

Bonds

The rally from the May low in bonds has carried about as far as it can if indeed it is a second or “b” wave from the March peak. Any further strength would most likely eliminate that count and allow for the possibility of a test or modest move above the March high. The rally from May is best counted as corrective on the daily charts. However, it is possible that the bonds have traced out a couple of 1’s and 2’s complete with the second wave being irregulars. And if that is correct we could be on the verge of a third of a third. If this does turn out to be correct we would view it as a “c” wave from either January of 2000 or October of 2000. The daily range oscillators are approaching overbought levels but have confirmed price. The daily trend oscillators have turned up and short-term momentum is positive The weekly range oscillators are in an up-trend and not yet overbought. The weekly trend oscillators are positive and medium-term momentum is also positive. Market vane reported 59% bulls, which is negative while Consensus Inc was 46% and that is neutral. The commitment of traders report is showing that commercial hedgers have moved to a fairly substantial net short position while the small speculator is still net long. Sentiment is weakening but not yet excessively negative.  Support: 103 18/32 +/- 3/32. 103 +/- 5/32, 102 03/32 +/- 8/32. Resistance: 1`05 25/32 +/- 12/32, 107-107 16/32. Please note that all price calculations are based on a constant bond chart and not on any particular futures contract. Currently the constant bonds are running around 3/32 ahead of the December futures. The short-term pattern is getting extended but the indicators have improved and we are going to move from bearish to neutral. The medium-term is a difficult one. Sentiment has weakened and is on the negative side of the equation but is not et excessive. However, medium-term momentum is still positive. We are going to move back to bullish on the medium-term due to the position of the momentum indicators although it is important to note that the rally is getting rather long in the tooth.

XAU

The XAU has now recovered about 50% of the May 18 to July 16 decline and has done so in what is so far a clean three-wave rally on both the daily and weekly charts. The rally has also confirmed the decline from May to July as a three-wave pattern and the decline from May to July has confirmed the rally from October 2000 to May as a three. The daily chart shows that both waves up from July can be counted as five wave patterns. At present we are approaching the rally as a zig-zag, and as a “b” wave from May 18 that looks to be close to over. It is possible that the rally from October was it or it is also possible that the rally was only the “a” wave of a larger pattern but the corrective nature of the advance implies that it is most likely not the beginning of a new long-term bull market. It is either all or part of a fourth wave from either 1998 or 1996 or our long-term counts are off. Short-term momentum is approaching overbought levels but is still OK. Medium-term momentum remains negative. Some key sentiment indicators have turned quite negative. Market Vane and Consensus are in the high 50% low 60% zone and the asset level in the Rydex precious metals fund is at levels seen at or near tops. The rally is so far corrective and approaching important resistance. How we decline from here will be important. We are going to stay neutral on the short-term but with an eye towards the exit door. Medium-term we remain bearish.  Long-term we are neutral but with a bottoming bias.

Indicator review

Following the early August rally its been all downhill for the averages. The DJIA has lost 282 points or 2.68% in August while the S&P has fared even worse as it has lost nearly 50 points or 4%. The DJIA has managed to hold above its early July low on a closing basis but the S&P has closed at its lowest level since early April.  The S&P looks to have completed a break down below a potential triple bottom of the lows on July 11, July 24 and August 10. The NYSE Composite did manage to hold its July 24 low both on a close and print basis but by well less than a point but the overall chart patterns look poor especially the S&P. Price/ volume relationships have been plainly negative with volume contracting on rallies and expanding on declines. However we have not come close to anything even remotely resembling climactic from the volume arena at all. Breadth on the other hand has held up extremely well and in fact the past two weeks while the averages are lower both the daily and weekly A/D liens have moved higher. The high/low statistics have been mixed. The new highs have held up decently and have consistently outnumbered the new lows. However, the new highs have been dominated by closed end funds and preferred issues. The new lows have been expanding on the declines confirming price but are not setting the world on fire. The Russell 2000 has held up considerably better than the listed averages losing about 9 points or 1.84% in August so far. We are neutral short-term and negative medium-term. The Value-line has lost 34 points or 2.7%. The short-term is neutral and the medium-term picture is negative. The NASDAQ averages showed some small spots of relative strength in July but they were spotty and only very short-term in nature. They picked up in August where they had left off in July and following a few days of rally early in the month they have fared far worse than the DJIA or the S&P. The NASDAQ Composite is off 201 points or 9.7% and the NASDAQ 100 is off over167 points or 9.9%. They Composite closed at its lowest point since April 10 and the 100 since April 9. The short-term is oversold and could bounce so we will remain neutral.  We moved to negative on the medium-term last week and remain there. The DJTA, which did well in July has given it all back in August losing about 2% and is back below where it had ended June. We remain negative on the short-term. Medium-term it has been stuck in a trading range since early in the year. It is beginning to weaken but fro now we will remain neutral. The utility sector as measured by the DJUA and UTY indexes have remained weak but are nonetheless holding the late July low. We are neutral short-term and they are oversold so they could bounce a bit more.  The medium-term and especially the long-term remain bearish.

Momentum

The breadth oscillator has held up quite well and remains neutral. However, it never did generate any positive momentum on the rallies we did get. The volume oscillator is oversold and slightly on the positive side. The 3-day oscillator is neutral but weak. The McClellan oscillator is also neutral. It too never did generate any positive momentum but in spite of all the weakness in the averages it has not moved back below zero. The 10-day and open 10-day Arms have gone from deeply oversold to more deeply oversold and are at extreme readings seen on only a few occasions. The 5-day and 21-day Arms are also deeply oversold. However, the new 10-day Arms is only neutral and low neutral at that. Moreover, it just issued a sell signal on August 15 and while it looked like a poor signal on Thursday it did not look like such on Friday. The daily range oscillators are close to the high end of oversold. We are seeing a potential small bullish divergence on the S&P and its 13-day RSI. We also saw one on July 24 that led to a minor bounce. The daily trend oscillators are negative. The weekly breadth oscillators are neutral. The weekly range oscillators are neutral but approaching the upper end of oversold. The weekly trend oscillators are negative.

Sentiment

The sentiment composite improved a bit to +8 but is still borderline negative. Investors Intelligence last week reported a small rise in both bulls and bears. The former moved to 49% from 46% and the latter to 29.1 from 27%. However, the bull bear ratio is still quite negative and this key indicator remains extremely negative. Market Vane and Consensus Inc are at neutral levels. They are close to bullish but have a way to go before coming close to levels seen at the April low. The American Association of Individual Investors (AAII) reported more bears that bull last week but a very slight margin. This indicator is neutral. NYSE members were net sellers for the latest reporting period but the indicator remains bullish. The commitment of traders report on S&P futures remains extremely bearish as commercial hedgers are still heavily short while the small speculator is still heavily net long. The insider sell buy/buy ratio has eased a bit but is still at very bearish levels. The CBOE put to call ratio basis the 10-day moving average has moved up smartly and is close to but not quite at bullish levels. The OEX ratio remains low and near bearish levels. The Rydex ratios have improved and are slightly on the positive side.  The volatility indexes have moved up a bit but are no better than neutral. The VIX is close to where it was on July 24 and that led to a bounce, but it is not close to levels seen in April. The VXN meanwhile is not even close to where it was on July 24 and is in fact still very close to where it was at the May 22 high.

Summary and Conclusion

We use Elliott Wave analysis like we use the Arms index or the put to call ratio, and that is as another technical tool. We do not believe it to be the holy grail to the market and it is clear that there are a number of times when the wave counts are ambiguous at best. However, there are also times when Elliott can be very, very important. While the short-term patterns are not clear (but are getting more so). The long-term patterns are and the fact that the March-May (April for the NASDQ) was a clean three-wave or corrective advance under the wave principle is evidence enough that the bear market is far from over. As we have discussed, our long-term counts place the DJIA and the S&P in a large degree fourth wave from 1982. Since fourth waves can unfold as triangles or as big trading ranges, it is possible that this is how the market will progress from here and stay within a big trading range between the lows of March-April and the May high for quite a spell. However, at this point we do not think that is how it will play out but instead see the averages moving well below the March-April low. Whether that is going to occur now in the next few months or the final price low will occur closer to the next 4-year cycle low due next year we cannot be certain of. While there has been some improvement in some of the indicators they are more of a short- and medium-term nature and not long-term. A few important long-term indicators are not anywhere close to levels suggestive of a major price low. One such indicator is the percentage of NYSE stocks above their 200-day moving average. At every important low of the past 14 years (that is as far back as we have the data), this indicator has moved below 30% before a low of importance was seen, and in a number of cases it moved well below 30%. In 1987, 1990, 1994, 1998 and yes even in March-April of last year it moved below 30%. And if you don’t think that the April 2000 low was important it did lead to a new all time high in the NYSE Composite and the Value-line index. Currently this indicator sits at 60%. Now some may think this is an extremely bullish divergence but history suggests otherwise and no we do not think it is different this time. Another indicator we have discussed often times is the Investors Intelligence survey. The current reading of 29% bears is nowhere close to levels seen at important lows and only three weeks ago it reached its lowest level in over 3 years. Near the 1990 low it was over 50% and that lasted for weeks. In 1994 it got close to 60% and that occurred in a trading range of less than 10%. In 1998 it was not great but it did get to 46%, not great but not too bad, This indicator obviously needs a lot of work. Another indicator is the insider sell/buy ratio. This indicator, basis its 8-week moving average was at its highest and most negative level in over 8-years. Last for today is the VIX. We talk about this often on a short-term basis and even from that perspective it is only mildly positive if that. However, at both the October 1997 and October 1998 lows this key indicator moved close to the 60 level. At the April 2000 and March 2001 lows it got to 40. This lead to decent rallies and the NYSE Composite did make new highs in September of 2000 but the rallies were not close to the caliber of what we saw coming off the 1997 and 1998 low. At its current reading of  26.74 its not even back to where it was on July 24 or June 15 and those two led to only minor rallies the lows of which have already been taken out. Now it is possible that these indicators as well as overall psychology can be corrected with the market going sideways over the next year and in that case the triangle count may come to fruition. However, if a decline such as we have seen since May, where the S&P has already given back 68% of its March-May gain and the NDX 75% and these indicators are still at bearish levels it is our best guess that the trading range or triangle scenario is a long-shot and in our view at least at this time we se significantly lower prices ahead. The bottom line long-term is that the “Bear” still very much in control and that there is considerably more to go in both time and price. Two more points regarding the long-term. First the sentiment model currently stands at +8. This is a borderline neutral position. At the 1994 low it has moved to a +18 and at the 1998 low it had moved to +14. For the record it ranges from zero to +24 and moves above +14 are in the bullish camp. In late March it made it to +13 so we still need a lot of work. Last is the technical barometer. It has shown some minor improvement of late but at best it is only neutral. More importantly, we did get that –10 reading back in late April. That is the most negative reading we can get and far out paced the –8 seen in February of 1998. The barometer is designed, to forecast out 3 to 5 months out. July of 1998 was 5 months from the –8 in February. Interestingly, the May 22 high was only a month or so after the –10 reading but it is now 4 months out and the outside score is only at –2. This remains a big picture negative and we cannot reverse the signal until we get to at least a +4 reading on the outside position (for more on the barometer go to the web site). On a more short-term note late last weeks we did see some serious improvement in the CBOE put to call ratio with Friday’s one day reading well over 1.00. Readings over 1.00 tend to occur near short-term lows. Although there could be distortions due to the fact that last week was options expiration we have to view this as a plus. The Rydex ratios have also improved and are at levels seen near the July low. However, they are only short-term positive as they are not even close to where they were at the March-April low. The 10-day and open 10-day Arms are especially bullish with the open 10 at its most oversold level in years and the 10-day back to near where it was in March. However, at the March low we were fully oversold across the board with the McClellan oscillator below –200 and the breadth oscillator deeply oversold at –790. The only indicator that is oversold is the volume oscillator. Now some may suggest that in fact shows a bullish divergence and that may well be true. However, we do not view it that way especially given the long-term picture. Moreover, both the daily and weekly trend oscillators are bearish and both on fairly fresh signals. At the March low the weekly oscillators had been heading down for over 2 months and were deeply oversold. This is just not the case at this time. We do respect the Arms index and view it as bullish. However, we have two points in regards to the Arms. The first is that the open 10 is not too far from where it was on June 19 and all we got then was a continuation of the decline as the index worked back to neutral. The second point is that while we do certainly respect this indicator a great deal we might add the overall technical backdrop does not support what this indicator is saying. The bottom line is that on a very short-term basis we see enough to suggest that we are getting close to a rally but only on a short-term basis if that and we are going to remain neutral. The medium-term backdrop is at best neutral but with a clear negative bias. Officially we are going to remain neutral but with a negative bias.

QQQ and SPY traders are flat. Rydex switchers sold the 20% gold fund position per the Noon pacific hotline for a 4% loss. We are holding 10% position in Ursa. The morning hotline will be on at 7:15 AM Pacific

The chart above is a longer-term view of the VIX on a weekly basis back to 197. Note how high the indicator reached at the important lows of October of 1997 and 1998. At the April 2000 low it move to 40 and we got a decent rally lasting several months. At the March low it  got to 40 and we got a solid rally . The level we saw in May was not far from levels seen at important tops such as July 1998, and September of 2000.

The chart below is a shorter-term view of the VIX. Note that in spite of the sharp decline from early August and he fact that the S&P moved well below the July 24 low the VIX has not even moved back to the level seen on July 24 and that produced only a modest bounce. In fact it has been above 28 on three occasions since the May high and each one produced nothing more than a minor rally. At the August 2 low it had moved below where it was at the Febraury and May peaks.

Every important market low of the past 14 years has been accompanied with the percentage of stocks above their 200 day M.A. below 30% and in most cases well below 30%. This indicator has a long way to go before even getting close to that level.

The 8-week moving average of the insider sell/buy ratio hit its most negative reading in over 8 years. Although it has eased a bit the past couple of weeks it is still at levels associated with important tops not bottoms.  Speaking of bottoms note how positive it got at the 1994 and 1998 lows.