Secular Markets

[Packman]

I just finished reading "The Little Book of Stock Market Cycles", and also recently read "Investing in the Second Lost Decade". Both books talk about secular and cyclical stock market cycles. I was particularly interested in "the best six months" strategy. In short, it states that the best six months of the market (historically) are November through April. And it suggests investing in equity indexes in these months and moving to bonds and/or cash the other 6 months. ETF's would make this very simple.

I know this is (dreaded) market timing, but the historical results are impressive. I looked at my own portfolio, by month, for the last 5 years (I'm pretty much of a buy and hold guy) and it showed much better returns in the best six months versus the worst six months, except for one year, but that year was close.

This would seem to be an awful easy strategy. I know I've read the "sell in May and go away" theory, but I never knew it was this lopsided. So...does anyone use a strategy like this? I'm just looking for input. Thanks.

[MasterBlaster]

Note that the "Sell in May and go away, Buy back in October" strategy wouldn't have worked so well this year.

And I am afraid for what will happen before next May.

But, there is some statistical evidence that what you posted may work when averaged over many market cycles. At least it has worked (generally speaking) in the past.

But are you willing to test it ?

Journalists like to write about stuff like this...

http://www.forbes.com/sites/eamonnfi...d-go-away-not/

http://money.usnews.com/money/blogs/...her-bad-advice

[Packman]

Quote:

Originally Posted by MasterBlaster But are you willing to test it ?

I would not ever go all-in or all-out of equites. But I could see changing my stock allocation to say, 70% equites in the best six months and 30% equities in the worst six months.

By the way, I highly recommend The Little Book of Stock Market Cycles. It just was published this year and it's an easy read that makes you think. The first half of the book is better than the second half.

[haha]

I think if your assets are tax-deferred and in no load mutual funds with no penalty for early withdrawal, and your strategy is completely passive except for this calendar trading, why not try it?

My guess is that 80% of the people who do this will abandon it the first time they sell by this strategy and the market is 20% higher 6 months later. This is not a criticism of the strategy, more a fact about people.

Ha

[NW-Bound]

As I am neither an indexer, nor one who has timed the market with the method of "Sell in May", I have the following observation.

If it is true that statistics show that the "Sell in May" method works, that is not the same as guaranteeing that it will ALWAYS work every year. It only claims that over many years, it works. So, one must stick with it for many years to reap the benefits. See Ha's comment above.

Why would that not be the same as the statistics that show that index funds may lose to active funds occasionally, but will win over the longer term?

Why do people readily accept the 2nd premise, but reject the 1st one?

As I said, I am agnostic on the matter of the two above claims. I just wonder why one is more popular than the other, if both claims are supposedly based on actual statistics that can be readily verified.

[Tyro]

Did the book say anything about the presidential cycle, which, IIRC, states the last two years of a president's term are, on average, more profitable than the first two years (exceptions like the '80s noted)?

[MasterBlaster]

Quote:

Originally Posted by Tyro Did the book say anything about the presidential cycle, which, IIRC, states the last two years of a president's term are, on average, more profitable than the first two years (exceptions like the '80s noted)?

Isn't there also a Superbowl index that predicts a bull market when NFL teams win over AFL teams ?

- And pass the nachos !

[photoguy]

Sounds like a recipe for disaster to me caused by multiple hypothesis testing / publication bias. Here is a thought experiment: if you were to randomize the returns by months, how likely is it that you would find such a six month window if you tested of all of them?

I would never ever believe such a model for stock market returns unless the results were (1) reproducible across multiple countries / time-frames and (2) there was a logical explanation such as the risk story for value stocks.

[donheff]

Quote:

Originally Posted by haha My guess is that 80% of the people who do this will abandon it the first time they sell by this strategy and the market is 20% higher 6 months later. This is not a criticism of the strategy, more a fact about people. Ha

I have done OK holding tight through the downturns over the years but I have been tempted to go heavy on equities when I think we have hit a temporary low and/or to try a sell in May strategy. What keeps me from trying is the suspicion that I would pick a bad time/year to do it and then jump back in with bad timing as Ha predicts. I even thought of testing it with a small portion of my portfolio but that doesn't really make enough of a difference to be worth the effort.

[jebmke]

Quote:

Originally Posted by donheff I have done OK holding tight through the downturns over the years but I have been tempted to go heavy on equities when I think we have hit a temporary low and/or to try a sell in May strategy.

Many people don't even have the nerve to re-balance, much less do a tactical shift in their asset mix. Re-balancing equity buys takes care of any urge I have to "bulk up" on equity during a down turn.

[marko]

IIRC, Bill Gross said that more money is lost trying to time a market correction than is lost in the correction. I was proof of that for a while!

Buy in October, allocate yearly, hold on, RE several years later.

[audreyh1]

Quote:

Originally Posted by jebmke Many people don't even have the nerve to re-balance, much less do a tactical shift in their asset mix. Re-balancing equity buys takes care of any urge I have to "bulk up" on equity during a down turn.

Agreed! It's scary enough just bringing up the equity to a "normal" allocation after a major market correction. I can't imagine increasing the allocation under those circumstances. Last decade our bear markets haven't lasted that long, but what if they did last for many years......

[Texas Proud]

Quote:

Originally Posted by NW-Bound As I am neither an indexer, nor one who has timed the market with the method of "Sell in May", I have the following observation. If it is true that statistics show that the "Sell in May" method works, that is not the same as guaranteeing that it will ALWAYS work every year. It only claims that over many years, it works. So, one must stick with it for many years to reap the benefits. See Ha's comment above. Why would that not be the same as the statistics that show that index funds may lose to active funds occasionally, but will win over the longer term? Why do people readily accept the 2nd premise, but reject the 1st one? As I said, I am agnostic on the matter of the two above claims. I just wonder why one is more popular than the other, if both claims are supposedly based on actual statistics that can be readily verified.

Let me try and explain why I think people accept the 2nd and not the 1st...

The 2nd is based more on lower fees and trading costs... IOW, a well run index fund costs 10bp... where most active funds cost a full percent or so... that extra cost means they have to be a lot better year in and year out... people are just not that good at picking stocks.... even the pros... (see side note below)

As to the 1st... this is looking at historical returns and then picking a pattern... but does that pattern have a reason or is it just the way it was.... IOW, is this pattern based on some real thing that we just do not know, or did it just happen by accident and in the future another pattern is about to happen....


Side note.... I remember a long time ago when I got the Wall Street Journal they had a competition of pros picking stocks vs a dart (IIRC)... the darts did not do that bad... so, it seemed to show that pros were not that much better than any random pick... so lower costs make a big difference...


Looked it up:
Investor Home - The Wall Street Journal Dartboard Contest

[rbmrtn]

Quote:

Originally Posted by NW-Bound Why would that not be the same as the statistics that show that index funds may lose to active funds occasionally, but will win over the longer term? Why do people readily accept the 2nd premise, but reject the 1st one? As I said, I am agnostic on the matter of the two above claims. I just wonder why one is more popular than the other, if both claims are supposedly based on actual statistics that can be readily verified.

I am in the same camp. A timing strategy will not work all the time, know that going in. The season system does have some merit, Norman Fosback created it back in the '70s. couple of links to references

Fosback Seasonality Strategy « MarketSci Blog

Simple Tests of Sy Harding’s Seasonal Timing Strategy - CXO Advisory

Street Smart Report Online

I've used a system similar to Mebane Faber/Ivy Portfolios for a long time.

[rbmrtn]

Quote:

Originally Posted by Texas Proud As to the 1st... this is looking at historical returns and then picking a pattern... but does that pattern have a reason or is it just the way it was.... IOW, is this pattern based on some real thing that we just do not know, or did it just happen by accident and in the future another pattern is about to happen....

B&H is essentially the same thing. US markets tend to have an upward bias over time ( not so with other markets ). B&H invest based on previous market performance which has had an upward bias, works great as long as the trend continues and the market is not down when you need your assets.

[Tyro]

Quote:

Originally Posted by Texas Proud Side note.... I remember a long time ago when I got the Wall Street Journal they had a competition of pros picking stocks vs a dart (IIRC)... the darts did not do that bad... so, it seemed to show that pros were not that much better than any random pick... so lower costs make a big difference...

My recollection of that contest (or perhaps it was criticisms of the contest) is that shooting darts works during bull markets, but the pros allegedly did better during flat and bear markets, which I spoze may be part of the argument in favor of managed funds.

Regardless, what really counts (to buy & holders) is overall longterm performance through all types of markets, where the index funds apparently outperform managed funds.

Tyro

[Tyro]

Quote:

Originally Posted by audreyh1 Last decade our bear markets haven't lasted that long, but what if they did last for many years......

I've heard it argued that for the last decade we've been in a secular (prolonged) bear market (adjusted for inflation, and with intermittent rallies) that may continue for another 3-8 years (or more).*

*Not saying that I agree with that argument; I'm reading articles on both sides.

Tyro

[W2R]

Certain "experts" decide to analyze an overly short time series and think they see patterns such as the May/October pattern. Then they speculate as to definitive cause-and-effect relationships causing these patterns and with little to no scientific backing, present them as fact.

I don't want to seem negative or judgmental, but personally I will never make my own investment decisions based on these sketchy unproven hypotheses, because in my opinion such analyses of these patterns is very nicely described by Stevie Wonder:

Stevie Wonder - Superstition (Studio Version) - YouTube

[REWahoo]

Quote:

Originally Posted by W2R Personally I will never make my own investment decisions based on these sketchy unproven hypotheses...

Yep. Much wiser to do what many here do and rely on the tried-and-true, proven-beyond-any-shadow-of-doubt method of "Whee-cycling" your portfolio.

[NW-Bound]

The first thing to do with a hypothesis to see the evidence behind it, in this case the statistics of the seasonal return to see if it is true. Then, we would try to explain it. I guess the problem with finding the cause in this case is that it may have to do with human behavior. Humans are not that rational, yet their behavior may be predictable. For example, they buy high sell low. And do it again and again. Why? Just because...

As I said, I am agnostic to all this. Just wondering...