Stockmarket Cycles

The Cycles

On May 3rd of this year just a few weeks after the Value Line Composite arithmetic Index had made a new all-time high, and on the exact day when the daily advance/decline line had reached a new recovery high and the weekly advance/decline had reached a new all-time high, we published a front-page chart of the daily advance/decline line with Speed Resistance Lines a la Edson Gould. We noted at that time that the daily advance/decline line of the New York Stock Exchange had reached a potentially critical juncture. Because prices had just reached the 1/3 Speed Resistance Line, we said we should be looking for the possibility of resistance at that level if not a complete halt of the move. We are repeating that chart in today's newsletter, and updating the second chart in today's newsletter. God bless Speed Resistance Lines! If anyone is aware of another technique that pinpointed the top of the advance/decline line, we would love to hear about it. That very day, May 3rd, marked the exact top of the daily advance/decline line and, as you can see from the chart, it has since suffered a precipitous decline.

The year 2002 is scheduled to mark a market bottom of some significance. Based on longer-term cycles, we believe there is a good chance of seeing that bottom occur by October of this year. There is good reason to believe that most of the danger inherent in the current market will have been relieved by the end of October of this year. We say that because one of the most successful interpretations of the so-called "four-year cycle" is to simply count back to October of 1998, then continue back every fourth October. You have to go back 64 years or 16 resolutions of the four year October cycle before you find a bad October to be long stocks. In some cases a bottom had already been seen by October of the fourth year and in just a few cases (1994 for example), there was a slight additional risk below the low registered in October of the fourth year. In fact, you can go all the way back to 1926, a period of 76 years, and you will find only two four year Octobers where stock market risk over the next 12 to 24 months was present. It would be prudent, however, to point out that both of those vulnerable four year Octobers occurred in a deflationary environment, namely 1930 and 1938.

Another reason to expect the possibility of a bottom of some significance this year is the 20-year cycle which has provided some of the best buying opportunities of the past 80 years. The years 1922, 1942, 1962, and 1982 were all marvelous years to be buying stocks. The years 1942, 1962, and 1982 each registered stock market lows that were never seen again to the present day. We are not going on record that we feel the year 2002 will see the same resolution. In fact, we believe whatever low is registered in 2002 will almost surely be broken within the next two-four years. We do believe the cycles discussed above indicate there will be a substantial rally lasting at least six-nine months from whatever low is registered this year.

Technical Indicators

A few decades ago, we invented a technical indicator that we used to measure the 78-week cycle in the stock market. We will not go into the history of its development in today's newsletter. Those of you who are interested can read the introductory material regarding the Cycle Indicator (CI) (or the website under technical incicators), the Neutral Cycle Indicator (NCI) and the CI-NCI Ratio. This indicator has helped our market analysis on many occasions. The remarkable chart on page 3 depicts the daily closing price of both the Dow Jones Industrials and the CI-NCI Ratio over the past 40 years. When the market staged its spectacular rally last week, we were inspired to examine the CI NCI ratio for clues as to whether an important market bottom, a four-year cycle bottom if you will, had been formed. We were curious to see where the CI-NCI Ratio usually bottomed out as the four-year cycle was reaching its bottom. On the chart, we have circled every four-year cycle bottom since 1962 and used an arrow to point to the corresponding low reading on the CINCI Ratio. As you can clearly see, every previous four-year cycle bottom has seen a corresponding CI-NCI Ratio well below the neutral 1.000 level. The average reading at the bottom in the prior 10 cycle bottoms was 0.927. With the exception of 1978, no four-year cycle bottom was ever registered with a CI NCI ratio higher than 0.959.

When we saw these data, we immediately wondered what the reading was when the Dow Jones Industrial Average registered its low close of July 23rd. Was it possible that the four-year cycle bottom was being registered in July rather than in the fall which had historically been the more traditional season for a four-year cycle bottom of significance? It turns out the CI-NCI Ratio reading on July 23rd was a startling 1.008. In other words, there had actually been more advancing issues than declining issues on a daily basis over the prior nine months. We were aware of no important cycle bottom in the history of our data that occurred with a CI-NCI ratio reading above the neutral level of 1.000. For that reason, we not only believe the low that was seen last week was not an important one, we also believe the market is especially vulnerable over the next few months. We say that because a move to significantly lower levels on the CI-NCI ratio usually implies a move significantly lower in the popular averages. In fact, if we theorize that the stock market will not reach new all-time highs after the four-year cycle low due this year, we would also guess there is a good chance to see the CI-NCI ratio move to what we call "out of limit" readings. We discussed this in more detail in our introductory material, but an "out of limit" move to the downside would imply a reading lower than 0.910. There is no way to correlate movements in the CI NCI ratio with the magnitude of the equivalent stock market moves, but a movement of that magnitude from current levels on the CI-NCI ratio down to readings below 0.910 would imply very significant market damage. We say that, of course, with the understanding that very significant market damage has been seen already.

Subscribers are aware that Stockmarket Cycles is virtually 100% technically oriented. Once in a great while, however, some fundamental facts strike us as so compelling that we cannot resist sharing them with you. In the July 22nd, 2002 issue of Forbes magazine there is an article entitled, "The Great Stock Illusion," written by Daniel Fisher. We believe it is must reading for anyone who believes the propaganda that they can count on long-term stock market returns of 8%-12% based on prior history. The article discusses a recently published study done by Robert Arnott and Peter Bernstein and published in the Financial Analysts Journal. They deconstructed the long run historical return on stocks which has averaged 7% a year over the past two centuries. They discovered a large part of that return--close to 5%--came from dividend yields. A smaller portion, 1.4%, came from our earnings growth while a remaining sliver came from an expansion in price/earnings multiples. They argue that the 5% dividend yield is history with the current S&P 500 yield at only 1.5%. As to the bullish argument that dividends have gone out of style because companies are plowing their profits into share repurchases and internal growth, the authors say they studied what happened to earnings plowed back rather than paid out and found that companies did a miserable job of investing them. Fisher says that perhaps the most surprising statistic from the study is the meager 1.4 percent real growth in earnings per share, a number the authors calculated from data going back to 1871. There is much more to the study, but the authors (they are not Ivory Tower theorists, but actual money managers) conclude the real total return on stocks over the foreseeable future should be not much better than 3.5%. Furthermore, they conclude there is one sure-fire way that high historical returns could return p/e ratios (and thus stock price) could collapse, perhaps as much as 50%. If that happened, dividend yields would rise and investors could once again look forward to high real returns. As to those who argue long-term returns will be better than 3.5% because yields will go up, Arnott says they might be right but the process will be unpleasant.

Market Projections

In our last newsletter, we repeated the nominal four-year cycle projection originally given in our newsletter dated March 23rd, 2001. That projection called for a low of 879.63-1,067.37. We said that, theoretically, that was the lowest possible projection that could be given in this time period. When that projection was decisively penetrative on July 19th, it became apparent to us that something larger than a four-year cycle bottom was occurring. Over the past few weeks we have been experimenting with longer price projections. We will continue to appraise you of the results of that research, but currently we have made no definitive findings. It is our contention, however, that the convincing move below the nominal four-year projection by the S&P 500 and most of the other major market indexes was a clue that this market wants to go significantly lower. That fits in quite well with the research we provided concerning the CI-NCI Ratio earlier in this newsletter.

Mutual Funds

The nominal four-year cycle projections given in the last newsletter for the S&P 500 were reached almost perfectly on July 15th. For that reason, we recommended purchase of the Rydex Nova Fund on July 17th on a special telephone update. As fate would have, the convincing break below the nominal four-year projections caused a quick two-day collapse and by the time we exited two days later we had suffered a 9.8% loss. It was a painful loss because we had been well ahead for the year in a year when the great majority of investors had suffered very significant losses. Last week's trade brought us close to even for the year. Currently, both Rydex and Fidelity Select switchers are in 100% cash positions. All mutual-fund switchers should stay in daily contact with the telephone updates. We have two different specific model portfolios--one for Fidelity Select switchers and one for the Rydex Group switchers. How you distribute your own portfolio is up to you as an individual.