First, a few comments about the following data.
It does not include dividends so keep that in mind. During periods of high dividends, that can make up for some of the pain.
Further, the data has been cherry picked. I'd much rather start with a theory and then look to the data to find support for the theory. In this case I have started with the data and ended up with a theory. That's not a scientifically strong way to do it. That being said, I think the following data is decent trivia at the very least though.
Click on it for a larger image.
The first "compounded" field shows what would have happened if you had let the money ride. In other words, you bought stocks in every September of a year that ended with a 9 and sold them in every September of a year that ended with a 2. There are 24 years included and the second "compounded" field shows how you would have done in a typical year.
Note that even if you strip out the horrible returns of the Great Depression, you'd still have been doing poorly in future periods (with 1949 to 1952 being the only exception).
In sharp contrast, the following shows the historical returns of buying in years that end with 2 and selling in years that end with 9. That's 49 years in total.
Note that the typical return in a given year was -8.61% in the first chart and was 6.12% in the second chart. That's quite a difference. Don't just look at that though. Compare and contrast the individual periods.
Here's a wacky theory on what might be going on. Ever drive down the road and see that your odometer is about to rollover? What is it about all those 9's turning into 0's that makes it so interesting? In my opinion, it almost feels like "peak mileage". From 1999 to 2000, we had a lot of 9's turning into zeroes. We had a lot of other things turning into zeroes as well. Many dotcom companies are no longer with us.
So what does that mean now? Perhaps it is not best to be buying stocks in September of 2009 with the theory that selling in September of 2012 would be a good idea.
This is not investment advice. It is a wacky theory based on market timing trivia. In any event, I won't be buying stocks in September. In fact, I haven't owned stocks since 2004. I don't need any more wacky theories telling me not to buy. I've got plenty already. ;)
1969 was a particularly good year (not to invest). We put a man on the moon that summer. The Concorde was also introduced. We're done going to the moon it seems and the jet that could fly from London to New York in half the time was retired in 2003. Good times. Hey, at least we have a robust and resilient banking system using the very latest in modern financial innovation, or so I was told. That counts for something I guess. Sarcasm is a must these days, lest I go insane.
I made every attempt to provide accurate data in my tables but I cannot guarantee accuracy. I do provide the source data though so feel free to recreate it and/or at the very least check my work.
Source Data:
Yahoo: Dow Jones Industrial Average
St. Louis Fed: CPI-U
Remarkably, given the last 9 months or so, even the 2002-2009 period ain't that bad. Sure, it's taken the biggest 6 week rally since you know when to somewhat salvage things, and there's still 5 months to go so anything can happen, but hey if one plans on living 120/150 years or so, it seems a sure-thing strategy, that is, if started at birth.
ReplyDeleteFor the rest of us, we can lament the lack of foresight and sacrifice of our Great-grandparents, and/or decide to do some good for our distant progeny.
Patience is rewarded, but in the long-term we're all dead. :-)
AllanF,
ReplyDelete"Remarkably, given the last 9 months or so, even the 2002-2009 period ain't that bad."
If we continue to rally at this pace into September, it could even look rather decent.
That would also give this wacky theory some serious legs heading into the 2009-2012 period. Downlegs that is, lol.
Of course, this is just trivia to me. There are an infinite number of ways this mined data could make sense. The front runner is still coincidence in my opinion, but perhaps the longshot odometer theory has a chance too.
Here are some other longshots.
Maybe decades from now a secret society of numerologists ("New World Ordering", hahaha!) will be revealed and they simply can't stand the number nine. Who knows! Why did I pick September? Upon casual inspection, it seemed to do the most damage to the first table. What else is special about September? September is actually the 9th month! Good grief. I'm ready for a numerical conspiracy theory rubber room, lol.
Or perhaps this 10-based cyclical pattern is aimed directly at us as karma for not adopting the metric system (think "Nine Inch Nails").
All Your 10-Base Are Belong to Us!
http://www.youtube.com/watch?v=-41ve90rks8
And lastly, this wacky theory could eventually form its own cult and be self-fulfilling. You know, like technical analysis training courses provided by brokers whose sole desire is to have all their clients make more money (i.e., trade more often and churn more fees).
Self-fulfilling prophecy
http://en.wikipedia.org/wiki/Self-fulfilling_prophecy
"In other words, a prophecy declared as truth when it is actually false may sufficiently influence people, either through fear or logical confusion, so that their reactions ultimately fulfill the once-false prophecy."
If we can convince enough people that this wacky odometer theory has merit, then the first believers can make massive money off of it! The next believers can make some money off of it. And finally, the last to believe can be financially slaughtered by it. It's the American way! You know, like stock prices and housing prices only go up! Sigh.