A Closer Look At The Five Day Rule

 |  Includes: DIA, QQQ, SPY
by: Ploutos

In a letter to clients last week, Goldman Sachs (NYSE:GS) chairman Jim O’Neill took a pragmatic approach by only offering a soft forecast for 2012 equity market performance. His reasoning for this timid preview was that since 1950, when the first five accumulative trading days produced a positive return, stocks had traded up for the year “86.8% of the time.” Justifiably, Mr. O’Neill wanted to take a peek at the market performance of the first week before he hardened up his opinion. Since 2011 was not a year that fit this mold (the S&P’s first 5-day return of 1.1% eclipsed its annual return), I wanted to take a deeper dive into the numbers to see if the first week’s trading does in fact prove prescient about the directionality or magnitude of annual returns.

[click to enlarge]

Click to enlarge

Taking a broader data sample beginning at year-end 1927, the S&P 500 finished up in its first week of trading on 56 occasions.For those years in which the S&P began with positive performance, it finished up on the year 41 times (73.2%) and down only 15 times (26.8%). In years where the first week return was positive, the arithmetic average annual price return was 9.8% versus an average increase of 2.2% for years where the trading calendar began with a negative first week performance. Hopefully for equity bulls, this is a positive harbinger after the solid first week performance in 2012.

Only 23 years in the sample period saw a better first week performance than the outset of 2012, and only three of those years (1931, 1932, and 1946) saw a negative price move on the S&P 500 for the full year. The arithmetic annual mean return of those 23 years was 14.2%. The extension of first week returns into annual prognostications seems to hold for poor starts as well. Examining the 23 years in the sample period with the worst first week return saw full year losses in 10 of 23 years, and an average annual return of only 2.0%. The dreaded 2008 saw the worst start on its way to a -38% price return. The Wall Street Journal ran an article in its online Market Beat section on January 9th that stated that “the market has never fallen in a year when the first five days see gains of 1.8% or more.” The Journal must have been excluding the aforementioned three years which returned -47%, -14.8%, and -11.9% respectively.

Click to enlarge

While the sample size is statistically significant, efficient market proponents would argue that a sustainability of the predictive power of this first week performance could be arbitraged away. Behaviorists might counter with the oft-studied January effect that indicates higher returns in the first month of the calendar year. If a continuation of last week’s success is anomalous or a confirmation of a greater trend, market participants on the long side will welcome it.

Disclosure: I am long SPY.