Over the past three years, from the March 2003 low there have been three medium-term rallies. The first lasted one year from March 2003 to March 2004. The second lasted seven months from august 2004 to March 2005. The third began in late April and to this point it has lasted nearly four months. What is interesting about this is a time relationship where the second rally or wave lasted 58% of the time of the first wave. The post April rally will have lasted 58% of the time it took for the second wave four months after it began. That points to sometime this month but to be fair, since the rally started in late April this relationship would be more accurate at the end of August. We do not do much if any work with time relationships but we do find this one quite interesting as 58% is also very close to a Fibonacci 61.8%.
This also fits in well with the longer-term momentum picture as we have seen a series of lower peaks in a number of key momentum and internal indicators both on a daily and weekly basis. This includes all of our primary momentum indicators, the breadth and volume oscillators as well as both of the McClellan oscillators. Other momentum measure such as rates of change and RSI are showing the exact same pattern having made their post 2003 peaks in late 2003 to early 2004. The same pattern can be seen in the new fifty two week highs both on a daily and weekly basis. And to s one of the more important long-term measures of participation, the percentage of stocks above their 200 day moving averages, is showing the same pattern. We see this and the new highs not only on the NYSE data but within the S&P itself. This is classic tape action clearly showing not only a big loss of momentum but a contraction in participation on a longer term basis. This is classic behavior of the last wave or leg up in an aging bull market. We saw a similar pattern leading into the 1998 peak and to a bigger degree into the 2000 peak, and for the record nearly every top going back decades.
The one indicator that is not diverging is the A/D line. And herein lies the biggest bullish argument as we have heard, and rightly so that it is rare to see a price peak without a divergence in the A/D line. Historically we cannot argue with that view as it is rare indeed to see an important top with the A/D line confirming the major averages. However, it has happened and can happen again. At every important market top there is a hook and we think that the A/D line this time around may very well be that hook. Clearly, today's A/D line is not your A/D line of the past. It is dominated by a massive number of non equity issues such as closed end funds, I Shares, bond funds and preferred stocks. It is also dominated by small cap issues the darlings of the day. A dollar invested in small cap stocks can buy a lot more shares and a lot more issues than a dollar invested in a GE or IBM or PG and on and on. This along with decimalization in our view has skewed the A/D line to give a picture of false security and confidence. Never before did we have a bottom in the A/D line occur within one month of a new high in the NYSE composite and a near new high in the S&P in September of 2000. No, in fact the A/D line has historically bottomed after a price low in the averages and on one rare occasion, in 1949 right before but not two years in advance. This to us is another piece of evidence that the A/D line may very well be the hook discussed above, the hook hat keeps the bulls in at the peak and most likely a good ways down also.
We have just finished reading, and for the second time we might add a book titled "The Amazing Life of Jesse Livermore". This is a biography of the famous trader that many have named the greatest trader that ever lived. We highly recommend this book as there is a lot to learn from it in regards to the market and trading. One of the many Livermore pearls of wisdoms in this book was "If you cannot make money in the leading stocks you can not make money in the market. what Livermore meant by leading stocks was not the high flying speculative issues, even as far back as the early 1900's he pointed out that when these shares (the speculative issues) began to lead it was s sign of public participation and a furthe3r sign that the market was closing in on a top. Seems like nothing has changed in the last 100 years. What he meant by the leading stocks were the big blue chip stocks. He said that when they started to not make new highs on rallies that a problem was brewing. He would watch the tape much more closely for the inevitable signs of distribution. Granted, the tape is a lot bigger today but we think that this is as valid today as it was in 1907.
Last time we pointed out importance of GE and its relationship to the DJIA. When GE diverges a reversal either up or down is not far off. Last weeks the DJIA and the rest of the averages moved lower. The selling so far has not been overwhelming, although the selling pressure late last week did pick up but the GE signal looks to have been right again. The decline has further locked in divergences between the DJIA and the S&P, with the former so far failing miserably to confirm the S&P's move above its March high. This may seem like it is not very important but we will point out a few such divergences between the two that were indeed important. The first was January-March 2000 with the DJIA peaking in January and not confirming in March. The 1998 low saw a divergence with the S&P making its low in October and the DJIA in September. March 2002 saw the S&P make a new high with the DJIA and the most important market low of the last 30 years in 1974 with the S&P bottoming in October and the DJIA in December. The DJIA's own A/D line continues to diverge failing to confirm the new high in March but also with failing in July to move above its May peak and showing a number of small divergences in July. Obviously there is more than GE weakness within the big blue chip index.
Last time we touched on some of the divergences developing between the averages. This time we want to add to that list. One of the more important sectors in the market are the financial stocks and within that sector it includes the bank stocks. when Financials begin to fail and not participate it usually has long-term negative implications for the market. The banking sector may very well be the single most important sector in the market as it relates to the health of the economy. It rarely diverges with the S&P but when it does one should not ignore the implications considering that financials account for a large weighting of the S&P's overall capitalization. The bank index (BKX) left a divergence in place at the early March high failing to move above its December peak but with the post April rally it has left an even bigger divergence as it not only failed to move above its December peak but also its March peak. The last such divergence occurred in March 2000 with January 2000 which actually dates back to 1998. On a smaller degree we also have a minor divergence from the broker-dealer index, which last week failed to move above its July peak. This had been one of the stronger sectors during the April rally and although the divergence is not overly big the failure to confirm from one of the leading groups is not a very positive factor even if only for the short-term.
Last week the McClellan oscillator moved to its most oversold reading since April but it did so only two days off a new high in the S&P. An oversold reading such as we have now if it occurred after a decline has been in progress a while would have the potential to indicate a possible trading low or more. However, a deep oversold reading two days off a multi year high is not that kind of a signal but is instead better viewed as an initiation signal such as we saw in early January. In fact the pattern leading into the January 3 high from late December is eerily similar to the current pattern. We see the same sort of signal from the 3-day oscillator suggesting the decline is not a one or two day wonder but more likely the beginning of a short-term correction. Whether this will lead to a decline in par with January or more like late June we cannot say but we favor one in between the two and do not expect to see it evolve as strongly as what we see from March to April. At the same time the evidence is strong and gaining strength day by day that the market is in the very mature phase of the post 2002 cyclical bull run. There is the possibility that what we saw last week may very well have marked the peak of that run but we see it as more likely that this rally was the penultimate rally of this cyclical bull market with one last run to higher highs in some of the averages likely to unfold before this bull breathes its last.
