Over the past several months we have seen any number of wave counts with a number of them, make that the majority of them counting the post 2002 and 2003 rally as a five-wave pattern. That's fine if one wants to ignore the basic rules of the wave principle as laid out by R. N. Elliott over 60 years but it is not fine with us as we adhere strictly to the rules. and in so doing we are left with no choice but to count the structure as corrective not impulsive. This in layman terms means that the rally is counter trend, a correction of the previous decline and not a new bull market to carry the S&P to new highs. In fact, as we discussed in the Elliott weave section earlier, we see the S&P as being in the very late stages of this counter trend rally. within this structure we favor one last rally to a marginal new high above the early August high to complete the pattern from March 2003 and from October 2002. It is our view that Elliott is quite a valuable tool in helping to come up with a rational and logically thought out view of where one is in the market cycle and to help forecast the most logical path the market may take to get to its ultimate price levels. However, like all other forms of market analysis the wave principle has its flaws. it is not the "Holy Grail" of market forecasting nor for that matter is anything else we have come across over our 21 years in the industry. What we have found in working with the wave principle is that other areas of technical analysis such as momentum and sentiment are not only needed to confirm what the wave principle is suggesting through the price structure and Fibonacci price relationships but is in fact the key. In other words the wave structure although important to the bigger picture is not the most important part of the analysis.
We have rarely if ever made a change in our market posture based only on the wave counts. We require confirmation from the indicators. When the four areas we focus on, momentum, sentiment, price structure and Fibonacci relationships come together to support one another at a minor nut more so a major inflection point it often is a thing of beauty. We have seen that start to come together over the last several months and with each passing week the evidence is getting stronger and stronger. We have already concluded that from an Elliott perspective the S&P looks to be in the very late stages of a large and bearish pattern from March 2003. In Elliott terms this pattern is called a diagonal triangle, in technical terms it is a rising wedge. This is a bearish pattern in either discipline although under the wave principle there are more specific characteristics as to whether this is the correct pattern. In the current situation the pattern is in its very mature stage and there is even an outside chance that what we saw in early August was in fact the final wave of this pattern. The decline from that peak was not consistent with the end of a diagonal triangle and that does suggests that the last rally we are favoring may yet pass. However, we have seen these patterns start of with a decline not consistent with post diagonal characteristics but come on a little later in the initial decline.
The technical backdrop over the past two years is extremely consistent with the long-term wave structure. Each rally, and there has been three medium-term advances since March 2003, has seen a weaker peak in momentum. This is not based on an isolated indicator or two but on a wide array of measures from the McClellan oscillator to the number of new highs to indicators such as RSI and even our daily and weekly trend oscillators. Indicators such as the percentage of stocks above their 200 day moving average and other measures of participation such as the percent of stocks in positive trends are showing identical patterns of lower highs while the S&P has made a series of higher highs. We have two such divergences dating back to the March 2003 rally. This is a classic pattern in technical analysis, two such divergences and is akin to the three strikes and you are out. There are occasions when you get one last run in the indicators along with that final surge in the market that gives the impression that the indicators are breaking out. They move above the second lower peak and can sometimes even challenge the first divergent peak but end up failing. This is like a false breakout above the upper trend line of a rising channel that looks to be the next step in an acceleration but turns out to be the final blow off move. Whether we see that or not is any ones guess but this is more often the exception rather than the rule.
There are also other factors that develop as the cycle matures. The previous leadership begins to fail and breakdown. One of the areas of leadership during the post 2003 rally has been the financial stocks. They have not only been leaders but are also one of the sectors that tend to breakdown in the late stages of a rally. We discussed last time the failure by the bank index (BKX) to confirm the August high in the S&P. This was a warning sign to us and another pierce of solid evidence that our take on where we are within the price structure was correct. To this end, the BKX has continued to move lower taking out is July low and breaking a rising trend line drawn off the April and July low. The S&P has not yet come close to its respective trend line, which is currently near 1175. But the relative weakness in the BKX is not only short-term but long-term as well. There are also specific sectors and industry groups that are late cycle in nature, that come on late in the uptrend. We learned years ago from someone far wiser than us that energy stocks are one of these late cycle sectors. It is clear that energy stocks have been in a strong up trend going all the way back to the March 2003 low. in fact the XOI oil stock index is up over 150%, which has far outpaced all of the market averages. Yes it is true that this area is one of the major leaders of the post 2003 rally that has not yet broken down. But it is also support the idea that the entire rally is a counter trend, late cycle affair. In addition, the latest surge in the new highs this past week has been dominated by energy and energy related issues, another sign that the rally is in its mature phase.
The sentiment backdrop we have discussed ad nauseam but with an exception or two these indicators have hit levels consistent with wide acceptance and levels consistent with important market tops. A number of indicators have reached levels far in excess of where they stood in late 1999 early 2000. This includes the sentiment composite, such measures as the 4-week moving average of both Consensus Inc and Market Vane and the 10-week moving average of the AAII bulls divided by bulls + bears. The latter hit its highest level ever as did the 8-week insider sell/buy ratio, which hit is highest and most bearish level in its 33 year history. Some of these indicators such as the insiders have stayed persistently bearish for not only week but months and years. The sentiment combo has been above zero for 27 months. Sentiment as we painstakingly learned in 2003 is not a good timing tool but when combined with classic late cycle patterns in momentum and what is a clearly corrective and mature wave pattern is another story and does serve to add further support to the fact that the post March 2003 rally is extremely long in the tooth. The fact that the S&P is also approaching some very important Fibonacci resistance levels while this long-term deterioration is continuing adds even more to the argument that the S&P is closing in on an important long-term change in trend, that the cyclical bull market is near its end and that the secular bear marker that began in 2000 is close to reasserting itself.
As mentioned above we favor the idea that there is one last gasp rally left in this old and tiring bull before it breathes its last. The low last week and subsequent rally may or may not be the beginning of that last wave up. The rally did look to confirm a number of short-term momentum divergences as real. We also have positive readings on short term sentiment measures such as the CBOE put to call ratio and the rydex data. We have also seen some improvement in the AAII survey. We also see the possibility that the rally last week was nothing more than s correction of the August decline. The S&P did stop just below the 61.8% retracement of that decline and the usual positive seasonal bias of strength or better put a rally of some degree early in a new month has gotten off to a weak start. There is still a few days left to reverse this weak start to what is historically a very negative month for equities. As we have pointed out previously, a rally early in a month does not mean that the month will end in the black but a failure to rally early has a strong history of ending badly such as we saw in April and August.
About Larry Katz
