However, more important than whether or not December turns out to be a down month is the fact that monthly “seasonal” patterns are highly variable, and thus nearly useless for consistent market timing, even though – and especially because -- they often have highly-touted performance averages.

Election-cycle patterns are much bigger and result in statistically more consistent market-timing patterns, since they have fundamental drivers causing them.

Note in the accompanying charts that in nine of the past 11 four-year election cycles, the low in the mid-term Congressional election year was significantly below the low that marked the end of the previous year’s “weak six-month season” (where historically, on average, the stock market has produced no net gains) and start of the following “strong six-month season” (where historically, on average, the stock market has produced all of its net gains) in the post-Presidential election year.  For brevity, we’ll call this the lower-low pattern.

The lower-low decline below the previous year’s low – which for 2005 was 1136 for the S&P500 and 10,000 for the DJIA 30, both occurring on April 20^th – has averaged 13.5% for the past 11 election cycles.  That would result in declines of just over 20% both for the S&P 500 from its 1273 high today down to 1011 and for the DJIA 30 from its March 7^th high at 10,984 down to 8,741.  For other reasons, we fully expect the decline will be much worse.

To see this lower-low pattern, note the down-sloping red arrows on the charts from the post-Presidential election year high to the low in following mid-term (Congressional) election year.  The intervening, up-sloping green arrows are from the autumn (Sep-Nov) low at the start of the “strong season” to the spring (Mar-May) high for the spring season.  Compare the ending point of the down-sloping red arrows to the starting point of the up-sloping green arrows to see the nine of eleven times that subsequent years had lower lows.

The only two exceptions, 1998-99 and 1985-86, naturally were near the beginning and end of the last BAAC Supercycle Bull Market Period, when the stock market is secularly much stronger than average.

Pertinent to 2006, all six previous instances during BAAC Supercycle Bear Market Periods had lower next year lows, in the mid-term Congressional election year, averaging a whopping minus 21+%.  This was 50% worse than the minus 13.5% average for all 11 election cycles due to the flat underlying secular trend in the stock market.

The reason this lower-low pattern has more important forecasting power than, say, calendar month averages is because 81% (nine out of 11) and 100% (six out of six) odds for a lower low pattern are in huge contrast to the only 27% lower-low odds in the 33 other paired-year low-to-low patterns in the 11 Presidential election cycles during the past 44 years, where 24, or 73%, had higher lows.  See the numerical data and spreadsheet at the end of this report.

And more importantly, these extremely high 81% to 100% historically-based odds of lower lows in 2006 are supported by fundamental reasoning.  The fiscal policy-making actions promised in order to get elected typically result in both predictable and unexpected adverse consequences.  These problems typically fully emerge in the second year of the Presidential term at which time they are opportunistically criticized by the other political party going into the mid-term Congressional election.

In this vein, we demonstrated high historically-based odds of a coming economic recession in “The Political Cycle Is Bearish” section of the following August 2^nd article [SEE], which was confirmed by a host of coincident economic indicators presented in this October 21^st article [SEE].

Bob Bronson
December 6, 2005
Bronson Capital Markets Research
Editorial Archive