STOCKMARKET CYCLES
Prepared by Peter Eliades
The Cycles
In our newsletter dated July 18,1997, we showed you a chart from Ned Davis Research that used six different factors to compile a cycle composite picture for the year 1997. The composite cycle showed atop around the 30th week of the year, then a double top around the 33rd week. That double top was followed by a virtually uninterrupted decline into late October-early November. The high for the year so far on the Dow Industrials occurred in the 32nd week of the year. In that respect the cycle composite was remarkably accurate, but the projected sharp decline which was due to last until early November has failed to materialize, at least so far. The failure of the market to decline in what should be a weak period of seasonality could be used as a bullish argument.
We believe the market has at least one more seasonality hurdle over the next three weeks. We discussed that hurdle on our daily telephone update of September 29. Here is what we said, “We did some research this weekend concerning the seventh year within each decade and more specifically the current time period within the seventh year of each decade since 1900. In the nine previous seven years within each decade from 1907-1987 the Dow was down eight of nine times between September 30 and October 31. They were not insignificant declines. The average decline in the eight previous October declines was 9.9%. Even if you remove the 1987 crash month of October, the average decline for the remaining seven declines was a full 8%. An average decline from current levels would be 640-790 points by the end of October.”
The chart depicts what the Dow would do from current levels if it acted in a similar fashion as the average of prior Octobers within the seventh year of each decade. It is obvious that, so far, the Dow is easily outperforming the average of prior seventh year Octobers. If it fails to decline substantially over the next week or so, it will be another bullish argument for the market, at least short term.
On an arithmetic scale, the Dow is now moving in what can only be described as a parabolic rise. We thought it would be interesting to depict the Dow's parabolic rise on a percentage chart which is similar to a log chart. Arithmetic charts treat points gained as equal measures no matter what the scale is. It is important to remember, however, that 100 points gained from 700 to 800 represents a gain of 14.3%. A 100-point move from 7,000 to 7,100, however, represents only 1.4%. Any chart that showed a consistent percentage change from year to year would ultimately appear parabolic because the equal percentages each year would represent greater and greater point gains each year. We are showing you a hypothetical chart in this section to depict our point.
The next chart represents percentage moves and it is thus a true parabolic chart with an unsustainable rate of ascent. What the chart actually depicts is a 17-year rate of change for the Dow Industrials. In other words, rather than measuring the percentage change from year to year, each month the percentage change from the equivalent month 17 years before is determined. Notice that the parabolic line on the right side of the chart is moving virtually vertically. It is a virtual certainty that the rate of change line will move below the parabolic support line within the next few months. The chart strongly suggests that when that happens, a market pattern that has been in effect for over 30 years will be broken. It does not necessarily mean the bull market will end when that occurs, but there is a strong presumption it will. Notice the left side of the chart. The 17-year rate of change reached a peak around February 1966 and began a long slide into the 1982 low. You should also notice what the actual rate of change was in 1982. For a full year from September l981 to Septernber 1982, the 17-year rate of change was below zero. That means that over the 17-year periods from 1964 to 1981 and 1965 to 1982, an investment in the Dow, dividends excluded, showed a negative rate of return. Remember that when someone next extolls the virtues of long-term investing to you. We can promise you that in 1982 no one was doing studies on the advantages of long-term investing.
The fact that the market has, so far, not displayed the expected weakness in this period of unfavorable seasonality forces us to reassess the market's short- to intermediate-term outlook.
On Friday Octobef 3, there were 631 new 52-week highs for the day on the New York Stock Exchange. That was the highest raw number of 52-week highs in almost 15 years (October 1982). It is a remarkable number in the face of a market that continues to increase at a record rate of advance. Usually a final top in a bull market does not occur until several months or longer after the peak in daily 52-week highs. If that is going to be the scenario here, and we should say we are not at all sure it will be, there are probably two higher probability routes the market might take. It could simply continue higher, rising yet another 15-25% in an unbelievable blow-off (yes, yes we agree, we thought we already saw one of those), then simply implode in a giant reversal. The other possibility, and we believe the more likely of the two if the final market top is still several months away, is that we see a continuing decline into the November- December time period, then a final rally back to the old highs or somewhat higher on the Dow. That is the rally where we would see the deteriorating internals that usually accompany a final market top.
Some of the sentiment readings suggest that a decline rather than a rally is more likely from these price levels. We will discuss some of them further on in this section, but first we want to follow up with some insight on daily 52-week highs and lows. The method for determining new highs and lows for the year was changed in 1977. Prior to that time, prices for the first three months in any year were compared over a time window that went back to January of the preceding year. After March of each year, the time window shrunk to January of the current year. In March of the year the time window consisted of almost 15 months of data, but in April the time window was as small as 3-4 months. For this reason, it is difficult to get comparative data prior to1977. Since that date, prices have been compared over a time window of 52 weeks. What we do know is that, almost without exception, the stock market does not get into serious trouble until you begin to see an increase in 52-week lows. To give you some examples, let's examine some prior important market highs and the greatest number of new lows within three weeks prior to the top.
Date Of Top Max New Lows Within Prior 3 Weeks 7-17-90 58 2.9% 8-24-87 24 1.2 1-6-84 78 3.8 4-27-81 36 1.9 9-21-76 34 1.9 1-11-73 47 2.6 12-3-68 15 1.0 2-9-66 57 4.0
The third column above under the New Lows heading shows the percentage of the total issues traded that day on the New York Exchange represented by the new lows reading that accompanies it.
Although no important top within the past 30 years has occurred with fewer than 1% of issues traded as the maximum new daily lows over the preceding three weeks, it is interesting to note that at the most important top for secondary stocks since 1929, December 1968, there were several days of single digit 52-week lows leading to the 1968 high and there is no rule that says it cannot happen again. The history of the past 30 years, however, argues against it.
If you wish to compare the recent statistics with those given above, in the three weeks preceding the August 7 all-time high on the Dow, the greatest number of 52-week lows was 21. That represented only 0.6% of the issues traded that day on the New York Stock Exchange. History ways it would be a rarity if August 7 proves to be a final high.
We mentioned earlier that there are some compelling sentiment data that suggest even if the market is going higher, we should see a significant decline first. Here is a quote from the Money and Investing section of the Wall Street Journal dated October 8,1997. “There is wind at the back of both the stock and bond markets right now,” said XXX...Indeed, the bulls seem to become more daring by the day. Yesterday, 97,751 of the newly introduced options on the Dow Industrials changed hands on the [CBOE]. Of those, 90,417 were call options, whose buyers are betting the average will continue its rally.
According to Track Data, a New York Firm that monitors option-trading activity, investors were heavy buyers of options betting the Dow Average would break above 8200 and 8400 before the options expire worthless at the end of next week.”
We do not yet have sufficient data on Dow options to make a conclusive judgment on the usefulness of the put- call data there, but anytime you see a put-call ratio of 8%, it should be clear there is a scent of overwhelming bullishness in the air. It would be very rare if the market accommodated that kind of overwhelming bearishness.
There are some option data that have been around for a long time, the data from the Chicago Board of Options Exchange (CBOE) relating to the equity puts and calls traded there. Over the past week, the 10-day moving average of the put-call ratio there reached its lowest level in the past 9-10 years. It doesn't mean the market can't go higher here, but it would be truly surprising if the market advanced sharply in the light of such low put-call numbers.
Market Projections
On our daily telephone update of Friday, September 26, we said, “it would also be misleading if we didn't caution you that the same weekly projection chart this week generated an upside nominal 10- week projection to 8213.20 +/— 70 points on a theoretical intra-day basis.” The Dow had closed at 7922.18 that day, so we had to allow for a further rally of almost 200 points. Seven market days later, the Dow reached a theoretical high of 8218.34, within 5.1 Dow points of an exact projection. Since then, the Dow has fallen 256 intra-day points to today's theoretical intra-day low of 7980.87.
As for the projection we gave you in our last newsletter, a nominal 20-week projection down to 7164.20 +/— 120 points intra-day, it was invalidated, at least temporarily, last week. Again this week, however, the Dow fell below the nominal 20-week projection line (a 10-week forward offset of the weekly median). One could argue that, because last week's invalidation was by only a marginal amount, the projectioncould still be in effect, but it is not a strong argument. What we do know is there are no upside projections currently outstanding. Infact,at today's (October 10) close, the Dow gave a nominal l0-day closing price projection to between 7935 and 7950. If that projection is met on Monday or Tuesday of next week, it would give a closing price projection to between 7860 to 7880.
October 10, 1997 Peter Eliades
Stockmarket Cycles
P.O. Box 6873, Santa Rosa, California
COMMODITY FUTURES FORECAST WEEKLY REPORT |
DEAN WITTER FUTURES COMMENTARY
THE LATEST COMMITMENT OF TRADERS ANALYSIS FOR CURRENCY CONTRACTS
MYERS ON FUTURES |
NIKKO MARKET COMMENTS
SCHAEFFER'S RESEARCH REVIEW
FED PRICES SLIDE BACK TO $65
MONEY VERSUS DEBT TWO AGGREGATES DIVERGED
STOCKMARKET CYCLES |
WEEKLY OUTLOOK
INTEREST RATE WATCH |
THE ALLENDALE ADVISORY REPORT
NIKKO MARKET COMMENTS |
THE REAPER
Copyright 1997, by Consensus Inc. All American and Pan American rights Reserved. editor@consensus-inc.com
Hosted by:
One Crossroads Place
610 West Maple Ave, Suite WWW
Independence, MO 64050
(816) 252-4080
sysop@kcmo.com
