By the end of April of this year, there was a pleasant surprise for most market observers. The market was at new all-time highs — at least the Dow Jones Industrials, NYSE Composite and the Advance/Decline line were. For many market professionals, there was disbelief that the market had robustly exceeded its earlier peak just prior to the March collapse.
It is no secret that the market has been successfully climbing a wall of worry for almost a year. The reality, of course, is it has been doing it for more than four years. The numerous negatives have created a bearish sentiment which, of course, is technically bullish.
The stock market is at a very interesting point in its evolution. The bull market is aging and we are in that seasonal period where professional investors consider the old adage “sell in May and go away.” Technicians are well aware of the four-year cycle. Many argue that June 2006 was the cycle low. However, that was only an 8% decline and four- year cycle lows typically result in 15% to 20% corrections or more. We can assume the cycle is still in front of us. The crash of 1987 created a four-year cycle low that was actually due in 1986, which is very similar to now. In fact, as we approach the anniversary of the 1987 crash, ironically, we could be setting up for something similar. We have the 20-year cycle; the four-year cycle; and we are in a year ending in “7.” Historically, such years are frequently bearish or have large corrections.
Going back to January 1928, it is interesting to note that there have been only two times in stock market history that the Dow Jones Industrials have had 19 out of 21 trading days where the market was up (21 days is the average number of trading days in a month). The first was July 5, 1929 and the second was April 27, 2007. These periods exhibit extreme optimism and, therefore, usually follow through to the upside even after they initially correct. It may be significant to note that the two largest corrections in the 20th century were preceded by this type of exuberance.
Specifically, on July 3, 1929, the market had been up 18 out of the prior 20 days. This was followed by three more up days. On Oct. 3, 1968, the market registered its 18th up day out of 21 days. Exactly two months later in each instance, the market made its most significant tops of the prior century. The top in 1929 had a 90% correction into the depths of the Great Depression and the top in 1968 was a broad market top where the average stock fell more than 80% into the 1974 lows. The 1929 top took 25 years to recover and the 1968 top took 15 years to recover.
There are other periods where there were 18 of 20 or 21 days up. The key is that in 1929 and 1968 there was a series of days. The series of days of 18 or more up days began on July 3, 1929 and ended July 9. In 1968, it began on Oct. 3 and ended Oct. 7.
The time between those two tops was 39 years. Taking the most recent decline (1968) and adding 39 years brings you to 2007. As mentioned, the only other time other than July 5, 1929 when the Dow Industrials had 19 up days out of 21 days was April 27 of this year. Using the ‘two month’ rule, one could project a top in late June. Using the exact time between the two tops (Sept. 3, 1929 to Dec. 3, 1968) and projecting it forward takes us to March 3, 2008. Therefore, we have an expected top range from late June of this year to March 3, 2008.
“Getting toppy,” below, shows a symmetrical breakdown of the three advances of this bull market. The chart confirms my other technical work that projects a mid May top. The 52nd week would match the duration of the first leg advance — note that the chart is also in a fifth wave, implying an impending top. Therefore, my assumption would be for a top in 2007 and a second top around March 2008.
Garrett Jones is a partner with Peter Eliades of Stockmarket Cycles Management Inc. (www.stockmarketcycles.com), which is the general partner of The Plutus Fund. Jones is a recognized expert in gold, technical analysis and the long wave economic cycle. He can be reached at email@example.com.