Sell in September? Time for a Reality Check

Anyone paying attention to market news must know that September is the weakest month for stocks. Mark Hulbert, one of our favorite writers, calls it The Cruelest Month. He provides a table of returns, summarized as follows:

Notice from the table that in all but one of the last 11 decades, September was a below-average performer. In more than half the decades, in fact, the month's rank was dead last.

Why, given such an overwhelming record, would anyone question September's bad record? Because there is no good theory for why the month should be such an awful month for the stock market. And, without such an explanation, there's the distinct possibility that the statistical pattern is just a fluke.

As one can see, Hulbert is well aware of data mining. He mentions the popular "butter production in Bangladesh" example.

Hulbert next considers a number of hypotheses and invites readers to share their own ideas. His thought is that a good explanation would make the September story more convincing, although he shares the information that day traders do not wait for any hypothesis testing! The sophisticated audience of this site may not find that very convincing.

Hulbert's Market Watch colleague, Irwin Kellner, has a number of reasons for September weakness, including the possibility of a self-fulfilling prophecy.

Why Both Articles are Wrong

Here we are veteran debunkers of mythical market lore. Most people (including our employers when we started in the business) just want to see the data --- all of it! Traders all believe in patterns. The more data the better.

The idea of random results is lost on most, like the "day traders" Hulbert mentions. There are twelve months. There will be a distribution. Some will be good and others will be bad. Always.

What if there is a reason? A hypothesis does not really help. When you already know the outcome, any smart person can invent a compelling reason. New readers can revisit our discussion of this topic, where a group of very smart grad students were given a list of findings and asked to provide reasons. They did very well. Only after the class were they told that all of the relationships were reversed!

The scientific method works only when one begins with the hypothesis.

Let us try an experiment. Instead of taking the currently constituted months of September, put all of the individual trading days in a basket. (We know that this basket has a negative bias, but bear with us). From this basket we create trading months. From the days in the other months, we create comparison months.

This is an interesting approach to getting beyond the data mining issue. It is (unfortunately) not our idea but that of Andrew Moe. He writes as follows:

When comparing the months composing September to a random basket of days the results are random. Attempts to find seasons of non-randomness are frequently subject to data mining bias, as the same permutation test debunking the September drift is easily used to identify (falsely) statistically significant periods.

The study. Running a bootstrap permutation study on Dow data from 1960 to 2008 we estimate the empirical distribution of differences in monthly return between September and other months. We test the hypothesis that a random September is no more bearish than a composition of random days sampled with replacement. We find that the mean difference between populations is 0.0695%, yielding a p-value of 0.3612 – random.

Our Take

Here are four good reasons to ignore the September weakness articles.

1. This is a widely advertised theory. Even if you do not believe in completely efficient markets, one would expect some anticipation.
2. We have already had a decline of nearly 2%, exceeding the expected monthly decline on the first day. Should we now expect normal trading for the rest of the month?
3. The evidence shows that the September pattern is not a statistically significant deviation.
4. Other seasonal methods (Sell in May, Presidential Cycle) have not worked well in this time of turmoil.

Quite obviously we do not know whether stock prices will mover higher or lower during the rest of September. We do expect trading to reflect fundamental information about economic changes, as well as perceptions and trader lore.

Comments

[Denis Gould:]

Excellent note. Will be looking further to see if you give us some pointers on what to follow, rather than what not to follow!

Sep 02 03:06 AM

[Roger Knights:]

'Tember!

Sep 02 04:33 AM

[T.Geiger:]

This fits nicely with Taleb's book Fooled by Randomness.

www.amazon.com/Fooled-...

Sep 02 06:21 AM

[aarc:]

Lets look at the charts: Assuming we are now in the recovery mode since March 2009 after 18 months of punitive selloffs.

Compare September from October 2002 which was the bottom of 2000-2002 meltdown:

- Sept of 2003 started with a bang and ended with a whimper. Ending red but not much.

- Sept of 2004 was a winner.

- Sept of 2005 was a winner.

- Sept of 2006 was a huge winner

- Sept of 2007 was a winner.

So based on the statistics of recent past comparing apples to apples with September data in an upturn. September was a winner with 4 winners and only 1 small loser. One of the winners was huge which was Sept of 2006.

Since March 2009 the stock markets had been in the upturn rather than a downturn. Somebody might want to study how September performed during downturns. Compare orange to oranges.

One major similarity between Aug 2006 and Aug 2009 is that traders and investors alike were complaning to high heavens of the persistent rallies for months on ends with no significant pullback or correction. They simply got left behind and wanted to get in.

Well in Sept 2006; they were given 2 days of significant selloff right off the bat. After that, the markets simply kept going up forcing everybody to just throw caution to the wind and joined the rally.

That was the best ever 5 months back-to-back rally most traders and investors ever experienced in their more than 25 years of experience as one seasoned investor/trader had written in financial publications that I've read after January 2007.

Good luck to all. But be careful just in case.

Sep 02 07:52 AM

[Value Added:]

Jeff, good points. However, based on fundamentals, a downturn *this* September would not be a surprise. The markets are overbought. Much of the volume recently has been in a small number of stocks, suggesting manipulation. There are signs that the economy is turning around, true, but there are major weaknesses which should not be ignored. I remain bearish.

Sep 02 08:36 AM

[Tony Petroski:]

It may be that the results are just chance. I suspect there are reasons related to human nature and activity that produce non-random outcomes. As for this September, most traders probably have two recent Septembers in mind: 2001 and last year. If you touch the stove and get burned, is it wise to touch it again?

Sep 02 09:04 AM

[Schweizer:]

September will be far worse than average. Just watch.

Sep 02 09:20 AM

[enigmaman:]

Since March the markets have put the cart before the horse, that only works for so long, with 50% in gains, the easy money has already been made, economic recovery speculation is now giving way to reality, less worse, meeting forecasts wont cut it any longer, now there needs to be daily upside surprises and upwood revisions to keep the Air in this Balloon.

Sep 02 09:27 AM

[David White:]

I don't think you even need to worry about whether or not it is September. We have had a huge run up from the March lows (>50%). The earnings season for the S&P500 is over. The "too low" analysts predictions for earnings are no longer available for easy beats to drive the markets higher. We have not had revenue growth. We have still only had earnings declines (overall). We can probably expect stimulus actions to provide some "growth" push, but we have already seen that the automakers results were worse than expected in the face of Cash for Clunkers. Plus the automakers probably robbed future sales to achieve even those results. That does not make one optimistic that the rest of the economy will surge upward soon. The retail sales numbers are still negative. They have generally been worse than expected lately. According to Art Cashin (and many others) the fair value for the S&P500 is in the 850 to 880 range now. We are still far from that. With all of the weakness we have seen lately (job losses -298K vs. expected -250K), there is no reason to think the market should trade at a huge premium to fair value. There are huge downside risks still. The market simply needs to return to reasonableness. If September happens to be a traditionally bad month, that just makes it all the more likely that people's irrational exuberance will falter this month. The fundamentals are the basis of this likely market correction though. Eventually they tend to catch up with the markets.

Sep 02 09:41 AM

[buybigtires:]

I think the last week in September through early October, will give us one of the worst percentage declines in the history of the market. The past few days are harbingers of a massive correction that is yet to come. The market is overpriced, and bank closures are going to accelerate rapidly.

Get out of the market while you still can.

Sep 02 09:51 AM

[macsmart:]

If not in September, there's always October.

Sep 02 10:25 AM

[marklamb:]

this is a well written article, but it's difficult to compare this september to previous years.

looking purely at fundamental economic indicators, it's hard to justify the market climbing much higher without a correction. i wouldn't be surprised to see that happen this fall. in my opinion, the effects if this crisis will take years to work themselves out.

Sep 02 10:28 AM

[buybigtires:]

I believe one that marklamb's comment lines up with what Nicholas Nassim Taleb had to say in "The Black Swan". We can line up all the historical charts we want, but they really just give us a false sense of security that sets us up for an anomaly. I seem to recall a September about 8 years back that really took us by surprise.

Sep 02 10:45 AM

[Ricard:]

I seem to recall a September in the past 365 days that really took us by surprise.

On Sep 02 10:45 AM buybigtires wrote:

> I believe one that marklamb's comment lines up with what Nicholas
> Nassim Taleb had to say in "The Black Swan". We can line up all the
> historical charts we want, but they really just give us a false sense
> of security that sets us up for an anomaly. I seem to recall a September
> about 8 years back that really took us by surprise.

Sep 02 03:24 PM

[ebworthen:]

Three kinds of lies: lies, damn lies, and statistics.

After a 5 month run up of 50%, I would put the chances of September being a good month at 0.000064. That is, five months of 10% gains per month, or 1/5 of a chance that September will be the same: 5 months or 1/5 x 1/5 x 1/5 x 1/5 x 1/5.

The higher the market goes, the more likely it will fall. The degree of the fall will be roughly equal to or greater than the degree of the rise. Fall in the Fall. It's like a stack of grapefruit in the produce aisle and kids are pulling out the one's on the bottom.

Sep 02 03:40 PM

[rosey99:]

Interesting. The Inductive Trap is definitely at work here. The entire article is summed up in one line: "Traders all believe in patterns." I would say that people seem to project a pattern onto whatever data is available, and traders are simply people. Look at data long enough and a pattern will certainly emerge. As another commenter noted, Taleb describes this very well in "Fooled by Randomness." IMO the more shocking realization is that people do not learn from their inductive mistakes, i.e., they continue looking for patterns and correlations, and of course they always find them. Taleb's assertion that "that's how we are wired" is unsatisfactory for me, though I have yet to find an alternative that I am comfortable with.

Critical thinking, especially when it comes to one's own methodologies is extremely difficult. That's why so few do it at all, and an even smaller number do it well. Thanks for this piece, more evidence (as if we needed more!) that human beings are not rational actors, but rationalizing actors. Ain't an Achilles' Heel a bitch.

Sep 02 04:03 PM

[WAdem:]

Great insight, but isn't it just possible that due to the perception that September is a downer, many investors take this conventional wisdom into their trading consideration. In other words, are we simply creating a self-fulfilling prophecy? Another thought that comes to me is that traders coined the, "sell in May and go away", so they could close out their positions and take a long Summer break. The still do that which is why the recent rally has occurred on such light volume. It follows that when the pros come back from vacation, they pounce on the opportunities left by retail investors. Just a thought.

Sep 02 10:22 PM

[VALUEDIV:]

I will make my prediction for remainder of 2009 and 2010.
Given the lack of liquidity by ungrateful bailed out banks, the unemployment and resulting decreased disposible income, the foreclosure rate increasing through 2010, a derivative global float of $ 480 Trillion in the financial sector alone ( $700 Trillion total) that will be "called in for payment" as the underlying financial health is revealed, The situation is critical in September and for the next 24 months. If one must invest their cash, short commodities and financials.

Not being a doomsayer, but Realistic on VALUATION fundamentals of the market as a whole. Of course, the "day pump" will be played, but the overall TREND is DOWN.

Sep 03 03:03 AM