Several emailers asked the following related questions:
"How can we have a sell off in September if everyone is expecting it?"
Another variation of this was: "There’s been so much discussion about the 4 year cycle, isn’t it less likely to have any impact?"
We can look at this a couple of ways:
First, the answer to these questions might be found by looking at an analogous situation: The Dogs of the Dow. This mechanical investing strategy simply buys the 10 Dow stocks which have the highest dividend yield in equal dollar amounts, and then holding them for one year (a variation is buying the 5 lowest price of these 10).
Since the early 1970s, this method outperformed the Dow or S&P handily — 17.7% annually, versus 10 and 11% for the SPX and DJIA. In recent years, the Dogs seem to have lost their ability to substantially outperform. The reason why, as far as anyone can surmise, is that it got too popular. Perhaps prices on the Dow Dogs didn’t come down enough; maybe too many people piled into the stocks early. That process took several years to occur; Once enough people lose interest in it, it could become an outperforming strategy again.
Secondly, I wonder if we are seeing an abbreviated version of that these days with many in vestors approach to some of the bearish cyclicality.
The four year cycle and the seasonal weakness of September have become fairly well known recently. But I do not see them as being as widely believed or acted upon. We had a few recent years where "Sell in May" did not work. And despite its track record, the four year cycle is hardly believed by the bulk of investors. That makes it less likely to fail ala Dogs of the Dow — i.e., due to popular enthusuiasm.
From a sentiment perspective, it may have already happened.
Consider the May sell off — it might have been in anticpation of the 4 year cycle or the September October weakness. That led to an oversold condition, and the rally since the June 13th lows. Since so many people thought EVERYONE ELSE already knows about September/October, it could have been already discounted in market prices.
Hence, many people already played the "Everyone is bearish about Sept/Oct, therefore it is built into in the prices" game. The early selloff and recovery could have already reflected the realization and subsequent dismissal of concerns about these issues.
As you can see, guessing what the dominant sentiment is rather difficult proces. I prefer to stock with measures of sentiment reflected in actual buy and sell transactions. These are my favorites examples:
Put Call ratio
Dow/Nasdaq ratio (both price and volume)
Mutual fund cash levels
Stock and Bond Fund inflow/outflow
None of these are flawless, but they are quantifiable, and reflect actual votes with dollars. I find that’s far more accurate than surveys or anecdotal sentiment discussions.