2007: ANOTHER OCTOBER CRASH?


David McMinn


An indicator of a possible emerging US October panic was the Moon sited near the autumn equinox point (180 E^o) at the time of a record peak in the two months commencing July 15. This occurred prior to the most dramatic October panics in the 20th century - 1929, 1987 and 1997. The DJIA, S&P500 & Nasdaq peaked on July 19, 2007 with the Moon again located near the autumn equinox point. This suggests a potentially serious situation may emerge, especially as October panics were also most likely to happen in 7 ended years (2007: A US Bear Market?). The Moon's location near the autumn equinox, as well as other Moon - Sun factors, are discussed in the ensuing text.

Please note: The annual one day (AOD) rise or fall is the biggest % one day movement in the year beginning March 1. The indexes used were the 12 Stock Average for 1885 to May 1896 and the Dow Jones Industrial Average (DJIA) since May 1896. The time assessed was 12 noon Eastern Standard Time on the day of the AOD fall, with no adjustment being made for daylight saving. Closing values of the various stock market indices have been used throughout the text unless otherwise stated. E^o is the abbreviation used to denote degrees on the ecliptical circle, where as A^o is used for the angular degrees between the Moon and Sun (lunar phase). A glossary is available at the end of the text for those unfamiliar with Moon - Sun terms.

Autumn Equinox Point

The great October crashes of 1929, 1987 & 1997 experienced a record high, with the Moon sited near the autumn equinox of 180 E^o (see Table 1). The Moon was also found in the same position during the actual panic (McMinn, 2006). For all 6 events of peaks and crashes, the Moon was located between 160 and 185 E^o, which would be very unlikely to occur by chance. The timing obviously has something to Moon - Sun cycles, but very little is known of the actual mathematics involved. On July 19, 2007, the DJIA and S&P500 experienced a record high, while the Nasdaq witnessed a post 2000 high. On this date, the Moon was located virtually on the autumnal equinox, a finding that may indicate another looming October panic.

Importantly, historic October panics have always happened a few days prior to a new Moon and around the full Moon (see Appendix 1). Assuming this repeats for 2007, the possible panic dates would be:
*        Wednesday October 24 (near the full Moon).
*        Monday October 08 (prior to the new Moon). October 07 would align better with the 1929, 1987 and 1997 events, but this would be a Sunday a non trading day.

The record highs in 1987 and 1997 were separated by an interval of 3634 days or 123.06 synodic months and was equivalent to a Triple Hepton eclipse cycle. Amazingly, this 3634 day interval was repeated between the record highs in 1997 and 2007.

Most Moon - Sun cycles aligned in integral and half integral numbers at the 3634 day interval, the exception being apogee months. This means that the Moon and Sun align in similar relative positions to the lunar nodes and the equinox points, based on the angles of 0^o and 180^o (1st and 2nd harmonics). Variations on this theme were to be found for the intervals between the 1847 & 1857 British/US October banking panics, as well as various DJIA AOD October falls (see Appendix 2).

Lunar phase for October 24, 2007 would align with October DJIA AOD falls in 1927, 1937 and 1989.

October AOD falls (=>3.60%) post 1910 were followed by an AOD rise a few days after the panic, the anomaly being the 1927 mini panic.

These 6 October AOD rises/falls appeared in a very curious 2-8-60 year grid pattern (McMinn, 2006), based on synodic months and tropical years/months.

Market Peaks

All September - October DJIA AOD falls =>3.60% are given in Table 2 for the post 1910 era. Lunar phase data for the prior year highs or record highs has also been presented in this listing. For the 12 AOD falls, the 10 pre crash peaks occurred with the lunar phase between 330 and 085 A^o, a range of 115^o. This compared with the expected frequency of 3.8. All 12 highs happened with lunar phase between 280 and 115 A^o, a range of 195^o (significant p < .01).

If the record peak occurred in the 1.25 months after September 04, the autumn crash emerged within a few days and a correction market persisted for only a few weeks. 

In contrast, the record peaks during the 1.75 months prior to September 03 experienced a major panic at least 45 days after the peak and a post crash low after the panic. Interestingly, an interval of about 55 days was important in the following:
*      1929 - between the record high and panic.
*      1987 - between the record high and panic.
*      1998 - between the record high and post crash low.

For the July 16, 1990 record peak, the AOD fall happened 21 days later on August 06, 1990 (-3.32%) and a market low on October 11, 1990. Neither interval between these events could be linked to Moon - Sun cycles. 

Four major peaks occurred earlier in the year (February 15 to May 31) prior to major DJIA AOD falls, none of which were record highs. These four tops were followed by protracted bear markets lasting from less than 12 months to more than three years.

The 1929 & 1987 October black days took place two tropical months after the record peak, while the comparable interval for the 1997 event was exactly three tropical months. For 2007, the intervals between the July 19 peak and October 08 would be 2.96 tropical months and for October 24 3.55 tropical months. These intervals of 3 and 3.5 tropical months could be accommodated in the 1st and 2nd harmonics of Moon - Sun tidal cycles. Integral and half integral numbers of Moon - Sun cycles were found to be valid throughout the various Moon - Sun analyses.

The 1929 & 1997 post crash lows happened 0.5 synodic months after the panic day, where as the comparable interval for 1987 was 1.5 synodic months.

Why the intervals were in tropical months pre crash and in synodic months post crash was unusual and remained a total mystery.

Autumn Highs

Highs took place in the 2.5 weeks ended October 10 before the 1927, 1929, 1955, 1987, 1989 & 1997 panics. They all happened with a similar ecliptical position for the Sun (180 - 200 E^o) and Moon (260 - 310 E^o), while lunar phase was between 065 - 120 A^o. If the cycle repeats, a similar peak prior to a possible 2007 crash could also be realised.

In 1929, 1987, 1997 & 2007, the DJIA lows between the record highs and the October highs occurred between 23 and 31 days after the record highs. The autumn high for 2007 has yet to be seen.

Other Interesting Intervals

As noted by a friend, there were 2745 days between the 1929 and 1937 highs. Curiously, this interval is closely repeated (2743 days) between the DJIA highs in January 14, 2000 and July 19, 2007. This gave integral and half integral numbers for synodic months and tropical months/years, but not nodical months/years or apogee months. A lunar eclipse cycle of 93 synodic months was not presented in the relevant literature. 

An interval of 11912 days was found between the major DJIA lows of April 28, 1942 and December 09, 1974. There was also 11910 days between the December 09, 1974 low and the July 19, 2007 high. This interval could not be linked to an established eclipse cycle and few of the Moon - Sun cycles aligned at integral and half integral numbers. 

Conclusions

The assumption has been that Moon - Sun cycles for the 1929, 1987 & 1997 panics will repeat in 2007 and so far this approach has held up well. 
*       The Moon was sited near the autumn equinox point for the record highs in 1929, 1987, 1997 & 2007.
*       The four peaks happened with the Sun sited 115 to 165 E^o, the Moon 160 to 180 E^o and lunar phase between 000 &  065 A^o.
*       There were intervals of exactly 3634 days between the record highs in 1987 & 1997, as well as 1997 & 2007.
*       A low was reached a few weeks after the record high as in 1929, 1987, 1997 & 2007.
If the parallelism continues, one would expect a high around early October followed by a major panic. As with all predictions - only time will tell.

A notable problem is that modern central banks are able to manipulate the markets much more effectively than in the past. Under the fiat monetary system, a central bank can create as much money as deemed necessary. In mid August 2007, the financial system was flooded with money and interest rates slashed to support shaky markets. The US Federal Reserve will intervene again to support US markets, if there is another bout of financial distress (the 'Greenspan put'). How this intervention distorts the influence of Moon - Sun cycles on financial activity can only be speculated upon.

© Copyright 2007. David McMinn. All rights reserved.


References

Kindleberger, Charles P. Manias, Panics & Crashes. John Wiley & Sons. First published 1978. Updated 1989. Revised 1996.
McMinn, David. Market Timing By The Moon & The Sun. Twin Palms Publishing. 2006 .

Glossary

Apogee is the point in the Moon's orbit greatest distance to the Earth.
Annual One Day rise or fall is the biggest one day % rise or fall in the year commencing March 1.
Ecliptic. The plane of the Earth's orbit around the Sun, which is inclined at 23.5^o to the plane of the Earth's equator.
Equinoxes are node points where the plane of the Earth’s equator cuts the ecliptic. At these points, the equatorial ascending node is where the Sun passes from below to above the celestial equator at 0 E^o (0 Aries - vernal or spring equinox at around 20 March) as viewed from the northern hemisphere. The equatorial descending node is where the Sun passes from above to below the celestial equator at 180 E^o (0 Libra - autumnal equinox at around 22 September).
Full Moon. Occurs when the Earth is positioned between the Sun and Moon (ie: the Moon - Sun angle is 180^o when viewed from Earth).
Inex separates two consecutive Saros series and is equal to 358 synodic months (28.95 tropical years). In the Inex period, the Sun repeats the same angle to the Moon with the north node 180 E^o on the opposite side of the ecliptic.
Lunar day is the period from moonrise to moonrise and equals 24.8412 hours.
Month, Apogee (or Anomalistic Month) is the time taken for the Moon to complete one cycle from perigee to perigee and equals 27.5546 days.
Month, Nodical (or Draconic Month). The time taken for the Moon to complete one cycle north node to north node and is equal to 27.2122 days.
Month, Synodic (or Lunar Month). The time taken for the Moon to complete one cycle new Moon to new Moon and is equal to 29.5306 days.
Month, Tropical. The time taken for the Moon to complete one 360^o cycle of the ecliptical  circle and equals 27.3216 days.
Moon's Orbital Plane. Plane of the Moon's orbit around the Earth, which is inclined by 5^o to the ecliptic plane.
New Moon. Occurs when the Moon passes between the Sun and Earth (ie: the Moon - Sun angle is 0^o when viewed from Earth).
Nutation Cycle. The time taken for the Moon's north node to complete one 360^o cycle retrograde (ie: clockwise) through the ecliptic circle and equals 18.6133 tropical years.
Node. In astronomy, these are two points (eg: equinoxes, lunar nodes, etc) where the orbit of a heavenly body cuts an astronomical plane or where two astronomical planes intersect.
Node, Moon’s. The plane of the Moon's orbit is inclined at 5^o to the plane of the Earth's orbit around the Sun (the ecliptic). The lunar nodes are the points where these two planes intersect. The north or ascending node is where the Moon passes from below to above the ecliptic (South to North). The south or descending  node is where the Moon passes from above to below the ecliptic (North to South).
Nodical Cycles of the Sun and Moon. This covers two lunisolar cycles - the nodical year (Sun angle to the north node ) and the nodical month (Moon angle to the north node).
Perigee. The point in the Moon's orbit closest distance to the Earth. It takes 8.85 years to complete one cycle of the ecliptic circle in direct motion (anticlockwise).
Saros Cycle. This is the interval between two similar eclipses (every 223 lunar months or 6,585.32 days) when the relative angles between the Sun, Moon and Moon’s nodes repeat to within a degree.
Saros Cycle, Triple (or Exeligmos) Because of the extra one-third day in the Saros (6,585.32 days), an eclipse happens every 18.03 years approximately 120^o longitude further West on the surface of the Earth. After three Saros cycles of 54.09 years from the precise date of an eclipse, another occurs at about the same longitude.
Solstice. Times of the year when the Sun is furthest from the equator at 270 E^o (December 22) or 90 E^o (June 21).
Year, Lunar  is the year based on 12 synodic months and it is equal to 354.37 days. This is found in astronomy as the interval between the Saros cycle (223 synodic months) and Metonic cycle (235 synodic months).
Year, Nodical (or Eclipse Year) is the time taken for the Sun to complete one cycle north node to north node and equals 346.6201 days.
Year, Tropical. (or Solar Year) Time taken for the Sun to complete one cycle of the ecliptic circle from spring equinox to spring equinox and is equal to 365.2422 days.


Appendix 1

OCTOBER PANICS & LUNAR PHASE

October panics always occurred a few days prior to a new Moon and around the full Moon. This has been a consistent trend over the past 200 years.

6 major US & British October panics were listed by Kindleberger (Appendix B, 1996).

Since 1896, there have been 10 DJIA AOD falls (=>3.60%) that took place in October. (NB: The annual one day (AOD) fall is the greatest % one day decline in the year commencing March 1.)

Combining these two lists gives 12 events, all of which have lunar phase between:

*    150 & 205 A^o, 1847, 1897, 1907, 1927, 1937, 1989. Around the full Moon.

*    315 & 350 A^o, 1799, 1857, 1903, 1929, 1987, 1997. Prior to a new Moon.

No exceptions. Remarkably, 8 of the 12 panics happened in 7 ended years, where as 1.2 could have been expected by chance. All 12 years occurred in years ended in odd numbers: 3, 7, 9.

The biggest one day falls (=>3.60%) in the month of October were also considered, in addition to the DJIA AOD falls noted previously. This gave a small listing of only five extra dates in 2000, as well as the early 1930’s, of which two did not fall in the segments 150 - 205 A^o and 315 - 350 A^o noted previously. The anomalies happened in 1931 (281 A^o) and 1932 (074 A^o).