wmeubank@ocp.kcmo.com
STOCKMARKET CYCLES
Prepared by Peter Eliades' Stockmarket Cycles
The Cycles
What an exciting time period we are facing. Our research with the Coppock Curve suggests we are at or fast approaching a market peak of significance. In our last newsletter, we used the prior three Coppock Curve Killer Wave signals to attempt to estimate a time period when this market will reach a final price peak. Our conclusion was that a market peak should be reached by June 25, 1997 under average conditions or, if the longest prior resolution was approached this time, it would allow for a market peak as late as August 12-September 12 of this year.
Are there any cycles or turning point periods that could guide us in making a more specific estimate of a final top? We believe there are. In our newsletter dated March 14, 1997, we discussed a pattern of approximately 1157 calendar days. Without going through the analysis again, suffice it to say that we concluded the average pattern was 1165.5 calendar days with a tentative resolution of April 10, 1997. As it turned out, a closing low was seen on April 11, 1997. We had thought the pattern might well mark a final market top, but it is obvious now that a low of some importance was seen.
Since April 11, the Dow rallied 15.5% on a closing basis to the close on May 27.
But watch this! If we insist on more consistency for this rhythmic pattern, we come up with different resolutions later in the pattern that seem to be even more satisfying. The chart in this section shows a pattern which has had excellent consistency since 1981 and has pin-pointed four of the most emotional and dramatic market turning points of the past 13 years. The last four resolutions have all been bottoms so we should be careful with our current interpretation. With that in mind, you should know that the next resolution is due June 23,1997. As you can see from the chart the average resolution period is 1176 calendar days or 813.7 market days. No resolution has been longer than 1185 calendar days. That implies that the resolution this time around should occur no later than April 4, 1994 (the last resolution) plus 1185 calendar days or July 2, 1997.
Let's add another factor to the timing equation. We went back and examined the most important market tops of the century with emphasis on the tops that were signalled by the Coppock Curve Killer Wave formation, namely December 1968, January 1973, and August 1987. We were curious to see if there was a consistency in the time of the final leg up that led to the major tops. We discovered there was indeed an impressive consistency. We used the major tops of 1929, 1937, 1968, 1973, and 1987 to measure the time span of the final rallies leading to the tops. The results were: 1929--78 trading days, 1937--63 trading days, 1968--65 trading days, 1973--57 trading days, 1987--67 trading days. The average of the five resolutions is 66 trading days and the extremes were only nine days (1973) and 12 days (1929) from the average. This implies that if we are forming a major top and if it is finalized with a typical rally, we should see a final top around July 17 with a range of plus 12 days (1929) or minus nine days (1973). This points to a time window between July 3 and August 4 1997. Note that the beginning of the time window is but one day away from the end of the time window given for the 1176-calendar-day cycle analyzed, namely July 2, 1997. It leads us to believe that the period between June 23 and July 17 could see a dramatic end to the record-breaking bull market, with a focus on July 2-3.
If we carry the study a little further and examine some other tops that were not quite as important we come up with these totals for the lengths of the last leg up: 1919--50 days, 1916--89 days, 1946- -77 days, 1961--55 days (to closing high), 1966--46 days, 1981--58 days. This time although the average remains consistent with our initial results at 62.5 trading days, the variations are greater with extremes of 46 trading days and 89 trading days. It increases our confidence that if we are facing a final top, we should focus on the time period 62-66 trading days beyond the April 12 intra-day low on the Dow as a focus point and work with a time window of 2-3 weeks on either side. The focus window would be July 11-July 17, but prior extremes would allow for a final top as soon as 46 days (June 18) and as late as 89 days (August 19).
We have to be careful when we deal with technical indicators that were devised by other technicians. One such indicator is the McClellan Oscillator devised by Sherm and Marian McClellan and currently being espoused by their talented son Tom McClellan who edits the McClellan Market Report. Tom is currently in Timer Digest's top 10 timers so he must be doing something right.
We believe the McClellan Oscillator is one of the best technical tools available to the technician. It is an overbought/oversold oscillator determined simply by using the differential between advances and declines of the New York Stock Exchange. The oscillator is the numerical difference between two exponential moving averages.
We have maintained for some time now that in order to make accurate historical comparisons between oscillator readings today and oscillator readings going back in time, the McClellan Oscillator and its companion tool, the McClellan Summation Index must be adjusted for the huge increase in the number of issues traded on the New York Stock Exchange over the years. For example, when the McClellan Oscillator moved to a reading of +236 on May 5, 1997, it was the highest raw reading in the history of the oscillator. We keep "adjusted McClellan" readings based on advance/decline ratios rather than advance/decline differentials and we believe the use of ratios allows accurate comparisons with other historic readings on the oscillator. The reading on our "adjusted McClellan" on May 5 was +44.53, a high reading indeed, but nowhere near a record high.
Looking at the data back to January of 1962, the May 5 reading of +44.53 was only the 115th highest reading. To show you what a difference the greater number of issues traded on the New York Exchange makes in the calculations, an almost identical reading of +44.55 occurred on November 21, 1962. At that time the"normal" McClellan Oscillator was at a reading of +96.7. On July 19, 1973, a very similar reading of +44.59 occurred on the adjusted McClellan. The normal McClellan reading was 130.49.
The main reason for mentioning the differences between an adjusted McClellan and the normal McClellan Oscillator is that we are fast approaching a reading of +3,000 on the McClellan Summation Index. As this section is being written on Thursday evening, June 5, the Summation Index has closed at 2845.5. As a general rule, whenever the Summation Index moves above 3000, it is an indication of outstanding upside momentum and it has meant we were in a market that has minor downside risk over the next twelve months. We also keep an "adjusted Summation" Index and as a general guideline, adjusted summation readings of above 1,000 correspond with normal summation readings above 3,000. For example the only time periods when the normal McClellan Summation has ever been above 3,000 have been: January-February 1971, February 1975, January-March 1976, October-November 1982, February 1985, March 1986, February-March 1991.
Since 1968, the only time periods our adjusted McClellan has been above 1,000 have been identical to these dates with the following minor additions: March 1975, December 1976, January 1977, February 1986, April 1986.
The only true additions were December 1976 and January 1977 because the other dates appeared in the same time periods as the +3,000 normal summation readings but simply started a few days earlier or extended a few days later.
The big point to be made from this somewhat esoteric discussion is that if the normal summation goes above 3,000 in the time period but the adjusted summation does not go above 1,000, we will not consider it to be sufficient momentum to decree a continuing bull market for at least another year. If our adjusted summation reaches above 1,000, on the other hand, it will deal a blow to the super bearish case. Today our adjusted McClellan Summation Index closed at +672.
Here is an interesting sidelight. It was announced today that the unemployment rate reached its lowest level in over 23 years. If you are a fundamentalist, you might consider that great news. As a technician we consider it just another factor in the good news that accompanies major market tops. Are you curious where the market was and what it did the last time the unemployment rate was this low? It was in October 1973 during the first year of the second term of a president who was in trouble because of some "water" something or other. Here is the chart of that period beginning before the October 1973 high and continuing on for the next 12-14 months. Incidentally, the rally in September and October 1973 is one of those rallies that showed great upside momentum but failed to foretell a continuing bull market. In fact just the opposite occurred. Even though the "Adjusted" McClellan Oscillator reached a reading of +46.59 on September 26,1973, a reading even higher than the +44.53 seen on May of this year, and the adjusted Summation Index reached a reading of +929 on October 12, 1973, the Dow reached a peak on October 26, 1973 and proceeded to decline over 40% in the next twelve months. There are additional similarities to the October 1973 backdrop. In a Barron's column dated October 29, 1973 (the arrow on the chart points to the week that includes that date), Alan Abelson wrote: "...There are no signs of the beginning of anything terminal. An awful wad of dough has been squirreled away in short-term paper, and, as rates come down, the attraction vis-a-vis stocks is bound to diminish, to the benefit of equities. Moreover, the amount of professional skepticism around remains reassuringly large...latest survey of advisory services shows a precipitous drop in the number of bulls, to 39% of the total, from 50% the previous week and 57.6% two weeks earlier. (The pro portion of bears, 28.8% in the latest compilation, has held comparatively steady.) Bull moves rarely end with most seers looking downward." The column appeared the very day of the high before the 40% plus decline.
Back to the present, the latest Investors Intelligence report showed 49.6% bulls and 28.1% bears. The percentage of bears is the smallest in five years with the exception of one week in late January as the OTC market was making a peak before falling almost 15% on an intra-day basis.
In early May, we noted that substantially higher nominal 20-week projections had been generated. On our update of May 2, we noted the Dow had a nominal 20-week price projection as high as 7680 on an intra-day print basis and the S&P 500 cash index had an equivalent projection as high as 887.84 intra-day. If this is the true blow-off part of the parabolic rise on the averages, it is possible to see prices even higher than that, but it should be clear now that this is a terminal move.
June 6, 1997
Peter Eliades Stockmarket Cycles
P.O. Box 6873,
Santa Rosa, California
Relocated 06-13-97
Last updated on 06-13-97
Hosted by:
One Crossroads Place
610 West Maple Ave, Suite WWW
Independence, MO 64050
(816) 252-4080
sysop@kcmo.com