If you would like to receive our work more frequently and timely then read and respond to “A note to visitors“ in the left-hand panel of my financialsense.com archive page.  See the performance results at the bottom of this copy of our December 19 e-mail to our private e-mail list.

From: Bob Bronson [mailto:Bob@Bronsons.com]
Sent: Tuesday, December 19, 2006 5:09 PM
To: Private E-mail List
Subject: Don’t Buy the Dip On the Usually Bullish Year-end Holiday Effect

Don’t Buy the Dip Based On the Usually Bullish Year-end Holiday Effect

Conventionally defined there are two year-end, primarily holiday-based rallies.  They are included in the loosely defined, but popularly referenced year end rally, which in turn, has the most significant bullish impact on the six-month strong season -- see the descriptive table below:

The first of these holiday-based rallies is the so-called Christmas Rally, which is defined as the last ten days of December.  This year it starts tomorrow, Wednesday December 20.  The second one is the so-called Santa Claus Rally, which is defined as last five trading days in December of each year plus the first two trading days in January.  This year it starts this Friday, December 22, or two days later than the Christmas Rally. 

The seven-day Santa Claus Rally is an interesting pattern that has repeated in 21 of the past 23 years, in which the percentage magnitude of the rally alternates between being bigger and smaller than the preceding year as illustrated in the following chart:

There are seven possible sequences of days and weekends for the week between the two holidays.  But the two different rally periods always have five trading days in common, so that the Santa Claus Rally is either one or two trading days longer than the Christmas Rally, as illustrated in the following table: 

We've also found it useful for short term timing purposes to study the combined rally patterns as the table illustrates below.  Since the two patterns together have an aggregate range of up to nine trading days, we analyze all nine days, which could be called the Nine-Day Year-end Holiday Rally, which yields more information than analyzing either rally alone.  The nine days are typically six days up and three days down.  Only once since 1949, have all seven days been up, and that was 1991.  Both 1963 and 1991 each had eight consecutive trading days up when the day before and after are included.

Incidentally, for those who are asking the question, the Nine-Day Year-end Holiday Rally shows weaker bigger-smaller pattern alternation than the Santa Claus Rally (15 rather than 21 out the past 23 years).  Unlike the seven-day Santa Claus Rally, the Nine-Day Year-end Holiday Rally is not a locked-in time cycle and thus it is subject to earlier-later pattern alternation as well as bigger-smaller pattern alternation.  It is subject to two degrees of pattern alternation for a couple of reasons.

First, the seven-day Santa Claus Rally includes the first two trading days in January that the Christmas Rally does not, hence the Santa Claus Rally picks up the triple bullish bias of the well-known month-end, quarter-end and year-end financial reporting and investor automatic reinvestment of those cash flow distributions.

Second, the Christmas Rally also adds up to three earlier trading days than the Santa Claus Rally.  The combined effect of not including the first two trading days of January – the second one of which is typically a very strong up day (12 times larger than the average trading day) -- plus up to two trading days prior to the Santa Claus Rally, which variously occur from one to three days before Christmas, explains the weaker bigger-smaller pattern alternation in the Nine-Day Year-end Holiday Rally. 

For the past 57 years including 1949, which covers almost two complete Supercycle Periods, as we both fundamentally and technically quantify them, the cumulative net gain for the Santa Claus Rally has been 1.53%, as reflected in the lower right corner of table above. This is a whopping five times bigger than the average gain of 0.31% during a typical seven-day run.  Therefore, it is clearly a non-random, bullishly-biased time cycle of note.

But since most every trader knows this seasonal anomaly, they game it by collectively jumping in since it worked last year, which enhances its effect, or they collectively sell it since it didn’t work last year, which reverses its effect from bullish to bearish.  This is what competitive observer-participant feedback does to preserve the “average” for the locked-in time cycle pattern, which keeps the market efficient.

So don’t forget to carefully consider the pattern reversing effect of bigger-smaller alternation that is most likely to be bearishly operative during the coming seven-day Santa Rally starting this Friday, December 22. 

Click to enlarge above chart

Click to enlarge above chart

Bob Bronson
Bronson Capital Market Research