September has a rap as a bad month for stocks.  However, losing performances in 1999-2003 are offset by strength in 1998 and 2009, leaving us with a net gain of about .4%/month since 1993.  Looking at the S&P 500 going back to 1983, the month has averaged closer to a 1% loss.  Going back to the turn of the century, September has easily been the weakest month of the year...for more discussion of this "September Effect", click here (and look towards the bottom of the page).

Looking at our seasonal Dow charts, it's interesting to note the historically low trading volume at the end of August and beginning of September.  Inevitably, some market analysts will look at this volume and bemoan the prospects for September (probably correctly), while failing to understand that this behavior is rather predictable and unexceptional.

There have been some very nice September performances since 1993 as well as the bad.  Choosing stocks that lost in a big way around the end of August 1998 resulted in gains around 30% in September 1998!  The strategy isn't the best general one for September, though.  More commonly, you'd be looking at stocks that fit the quarterly theme...buy stocks with high yearlong or long term gains, and sell long term losers.  That strategy worked well in 1993-96, as well as 1999, 2001, and 2002.

Picking recent losers (via, say, a loss on the last day of August) works well in the first half of the month, but doesn't result in big gains in the second half.

Although yearlong gainers continue to gain in the month, expensive stocks tend to drop.

The best September strategy has been to buy stocks that finish with a large high-close differential on the last trading day of August.  That has raked in better than 4% since 1993.  Another strong strategy is to buy medical, biotech, and healthcare stocks. Such stocks gained nicely in 1994, 95, 97, and 98, continuing a trend that begins in the second half of August (especially August 1994, 1999, and 2001).  These stocks have done well in both halves of the month.  Despite the negative market and the volatility associated with biotechs, you'd still walk away from September with gains over 3% following this strategy.  Regarding the biotechs, these gains are best attributed to the fact that a number of important biotechnology conferences, where positive clinical developments are sometimes released, occur around this time of year.  It might be interesting to try identify exactly which biotechs...big/small cap, cheap/expensive stock price, cancer vs. other maladies...seem to prosper most during this time frame.

Regarding healthcare stocks, we don't have a ready-made explanation for their gains.  Obviously, they're superior to biotechs from a risk-adjusted perspective.

Banks have also done well.

Computer-related issues, especially semiconductors, have been poor performers.  In given years, though, computer stocks have done nicely, including some Septembers where the general market suffered.  Avoiding transportation-related issues and basic materials wouldn't be a bad idea either.

"Industry momentum" doesn't seem to come into play in this month.  Buying stocks from industries that have been particularly strong or weak doesn't seem to aid in the quest for profits.

Buying stocks that have a history of strong performances in September could be of benefit.  Even after we removed biotech stocks entirely from our data, a tendency for historically strong stocks to prosper persisted.  Likewise, there's a tendency for historically weak stocks to suffer.  The following table shows some of the best and worst September performers on a historic basis:

Cutting the month into halves, the first half has been flat even with the disastrous effects of 9/11 on the market.  Retails show strength in this time slot, but reverse in the second half of the month.

Looking at our second half data, it has been awfully difficult to consistently make money in this period.  This is not to say big gains haven't been available in given years with given strategies, but just that last year's winning strategy could be this year's loser.

Speaking of 9/11/2001, though losses occurred across the board, you might be surprised which stock group actually did the best in resisting monster losses on a risk-adjusted basis...the most volatile stocks!  On the other hand, these volatile stocks indeed lost a great deal of real money in the second half of the month, trailing only stocks that had already suffered big losses over the month of August.  The lesson we gather here is that following a catastrophe of this scale, the best strategy isn't necessarily to run out and buy put options on the most volatile stocks you can find, since they might not suffer to the extent you'd expect.

Cheap stocks and low-cap stocks also resisted losses relatively well following 9/11.   Yearlong winners were also hurt when the market re-opened, but recovered nicely and actually finished the month as one of the groups that did best in the full month.

To look at the gains and losses in the trading session immediately following Sep 11 (Sep 17), go here.  Note that communication and scientific equipment makers held up relatively well.  Hotels and casinos took heavy losses, particularly after risk-adjustment.  Stocks that already had a bit of momentum, as measured by monthlong gain or our "momentum" indicator, also held up well in this trading session, while groups that were already suffering (e.g. stocks trading well below their short term averages) lost as much as 10% in the session.

Looking at the large-cap S&P 500 (as opposed to the Russell 3000, our usual source of stocks) with data going back to 1983, the themes above remain in focus.  Yearlong gainers have averaged about 4% gains per September, though the presence of several years where this group reversed means that going with this group every year since 1983 would only have resulted in a breakeven performance after the gains and losses were compounded (still, pretty good given the tendency toward losses in the month).  The cheapest 10% of S&P 500 stocks have fared well in the month.  Industry groups are not featured at all here...bear in mind that the rather stodgy S&P 500 holds only a few biotechs.

The S&P 500 data also re-emphasizes that late August winners and longterm losers should be avoided.  Stocks with a poor September history would also be better left untouched, while stocks with solid September histories could be considered purchase fodder.

************************

Our viewers should be aware of a couple "September effects" that have been cited in the media.  One is simply the fact that stocks tend to lose in September. 

Another, though, is a tendency for industries that gained nicely in August to continue to gain in the next 12 months.  Sam Stovall from Standard and Poor's notes that holding the 10 industries with the biggest gains in August over the next year resulted in average 16.9% gains versus 9.7% for the general market (as measured by the S&P 500) over the period 1970-2003.  We'd have named this the "August effect" ourselves...but we didn't make the initial observation.

Copyright © 2010 MarketSynopsis.com. All rights reserved.