It seems you have JavaScript disabled.

Ummm.. Yeah... I'm going to have to ask you to turn Javascript back on... Yeah... Thanks.

Will September Kill The Rally? at Trader’s Narrative

Will September Kill The Rally?

We’re getting a distinct message from the sentiment data that this rally, whether a bear market or cyclical bull market (or the first leg of a secular bull market) has overstayed its welcome.

If you have bearish leanings, then you wouldn’t mind comparing this recent 6 month old rally to ones that we saw in the aftermath of the great bear market that followed the 1929 top:

bear market rallies 1929-1932 compared to current Aug 2009
Source: Chart of the Day

This rally is now longer than almost all counter-trend rallies during the 1929-1932 bear market. It is only comparable to the biggest one (November 1929) which lasted 155 days and lifted the Dow by ~50%.

stock market gains from cyclical bull markets first 5 months and 6 to 10 months after investechIf, on the other hand, you have bullish leanings, then you would appreciate that we are following fairly closely the script provided by the aftermath of previous bear markets. Next up would be a shallow retracement and then a rather protracted trading range (click to see graph in previous link).

As well, looking back at previous cyclical bull markets since the 1940’s, we again seem to be following a familiar script. The data to the left is courtesy of InvesTech Research dividing each cyclical bull market into two phases. It is normal for the first 5 months of a rally to be robust while the following months are more lethargic. In fact, on average the returns in the first 5 months have been double those of the next 5 months. Which would again imply that going forward, we can not expect similar returns.

Don’t forget, we are also about to step into September - a most notorious month for stock market returns. Mark Twain famously wrote:

October: This is one of the peculiarly dangerous months to speculate in stocks in. The other are July, January, September, April, November, May, March, June, December, August, and February.

He was close, as he should have singled out September as the most dangerous time to be long. From 1896 to 2008, the average return for September has been -1.2%, the very worst of months. As well, in the more recent market history (20 and 50 years) the negative performance drag of September has been even more pronounced. Moving on, after September, the next worst month is February (probably because of profit taking after the January effect). And the best month? December, of course.

To illustrate just how damaging September is, imagine for a moment that you had a rich ancestor way back in 1900 who, for some masochistic reason, decided to invest the princely sum of $1000 in the Dow Jones but only during the month of September (and go to cash the rest of the time). Here’s the value of your inheritance over time:

1900 to 2009 investing just in September ouch
Source: Have a happy September. Or not.

Believe it or not, after more than 100 years of investment in the Dow, the original $1000 would be worth just $251 today. That’s a heart-stopping loss of -75% for over 100 years of one month investment. So obviously, doing the opposite would give you a leg up. But you may not believe just how much of an improvement you would squeeze out of just skipping one single month! Compared to a simple buy and hold strategy through thick and thin over the 100+ years, skipping September would give you 4 times the return!

Enjoyed this? Don't miss the next one, grab the feed  or 

                               subscribe through email:  

3 Responses to “Will September Kill The Rally?”  

  1. 1 Tyler

    Since 1900 September has posted a positive return 49%, as you point out, with an average return of -1.2%. When July and August are up, as they have been this year, September has posted positive gains 59% of the time in 39 instances. The average return for all 39 instances (including the down-months) was a relatively flat .09%. Thus, the probability and magnitude of a down September after a good July & August is slightly less pronounced.
    This was pointed out by Citi strategist, Tobias Levkovich.

  2. 2 wayne

    As we have noted, since 1950, September has been the weakest calendar month of the year, but normally there is a modest bullish bias the week (5 trading days) before Labor Day.

    The pre labor day rally is getting off on the wrong foot this week, as we are already down 2.5% as I type. If the week before Labor day is down, the prospect for the remainder of September gets worse. In the 20 years since 1950 when the SP 500 was down the week before Labor Day, the odds for the rest of September go to 7-13 with an average “additional” loss of 1.39%.

    Stay tuned

  3. 3 Babak

    Here’s a table of stats (thanks Wayne) since 1950’s for Sept’s performance

Leave a Reply