WHY A LOSING FIRST WEEK OF YEAR ISN'T GOOD -- A STUDY OF JANUARY'S PREDICTIVE VALUE -- YEARS ENDING IN FIVE ARE USUALLY GOOD -- BUT FIRST YEARS OF PRESIDENTIAL CYCLES AREN'T -- WHY JANUARY IS THE BEST MONTH TO TAKE PROFITS
THE JANUARY BAROMETER... You're going to be hearing a lot over the next few days and weeks about the importance of the month of January in determining stock market trends for the rest of the year. There are two January signals that I'd like to explain -- one having to do with the entire month and one having to do with the first week. Let's start with the January Barometer which holds that as the month of January goes, so goes the rest of the year. According to the Stock Trader's Almanac, the direction of the S&P 500 during January has done a excellent job of predicting the direction of the stock market for that year. Since 1950, the S&P 500 gained 35 times during January which was followed by 32 up years (91%). In the 19 Januarys when the S&P 500 fell, the market fell in 11 of those years (58%). There are two conclusions to be drawn from the data. One is that up Januarys have a better forecasting record than down Januarys. The other is that in most years that the market fell during January, the market had a bad year. A bad January doesn't ensure a bad year. But it tilts the percentages in that direction. The direction of the S&P during the first week of the year also has had predictive value.
FIRST WEEK OF JANUARY IS ALSO IMPORTANT ... According to the Almanac, "the first five trading days of January often serves as an excellent early warning system for the year as a whole". Here again, it's track in up years is better than in down years. In the 34 years since 1950 when the S&P 500 rose during the first week of January, the market had plus years 30 times (88%). In the 20 years when the S&P fell during the first week of January, the market declined that year 9 times (45%). So the track record during the first week of January isn't quite as good as it is for the entire month, but it's good enough to pay attention to. It looks like the market is going to end lower during the first five trading days of 2005.
WHY JANUARY IS A GOOD MONTH TO TAKE PROFITS ... I mentioned yesterday that January was a good month to take profits. That has to do with monthly seasonal patterns. The strongest three month span of the year runs from November through January. December is historically the year's best performer with January a close second. That makes January an ideal month to take some profits after a fourth quarter rally. After a down (or sideways) February, the market usually rallies into April after which the market weakens during May. That makes April another good month to take some profits. [April also ends the best six months of the year which usually starts in November. Hence the phrase "Sell in May and Go Away"]. A third good month of the year to take some profits is July after a summer rally. The best time of the year to put new money into the market is during the traditionally weak September/October period. The best seasonal month of the year to take some profits is January.
YEARS ENDING IN FIVE ARE ALWAYS UP ... You're going to be hearing a lot about the bullish history of market years ending in the number five. That's based on the Decennial Cycle which breaks market history into ten year periods. According to the Stock Trader's Almanac, the market has weak years ending in zero (remember 2000?) and seven (remember 1987?) and strong years ending in eight (like 1998) and five (like 1995 and 1975). The Dow Industrials haven't had a losing year ending in five since 1881. That extraordinary record is sure to give bullish analysts encouragement this year which ends in a five. There is one caveat regarding this year however. And that has to do with the four-year presidential cycle.
FIRST YEAR OF PRESIDENTIAL CYCLE IS THE WORST ... The four-year presidential cycle holds that the market usually bottoms every four years (the last four being 2002, 1998, and 1994). As a rule, the market and the economy do worse in the first two years of the four-year presidential cycle and better in the last two years. The best of the four years is the pre-election year (which was 2003) followed by the election year (which was 2004). The worst year of the president's term is usually the first year (which is this year). The reason for that poor record during the first year of new presidential term is that the government usually takes whatever strong (read unpopular) measures that it feels are needed to prop things up during the second half of the term (as elections approach). Along those lines, I find it interesting that Mr. Greenspan waited until after the last election to voice his concerns about our huge budget and trade deficits and to warn about higher interest rates. It's also interesting that the Fed minutes released this week for the first time voiced concern about the threat of inflation and the need for a more aggressive rate tightening cycle. More than any other factor, those minutes probably accounted for this week's poor market action. Then again, it's the first year of the cycle. What better time for some tougher measures? If the four-year cycle plays out again this time, the next major buying opportunity should take place in the autumn of 2006 (next year) in plenty of time for the next presidential election in 2008. The arrows on the S&P 500 show that the market tends to bottom every four years. The arrows fall in 2002, 1998, 1994, 1990, 1986, 1982, 1978, 1974, and 1966. The next bottom is due in 2006.
Chart 1
