How to play the fall market slump

By Jim Jubak

Sometimes a market trend is so strong, and so historically reliable, that you should just go with it. That's the case with the market's tendency to slump from late August well into October. Then the market, most years, rallies into the end of the year.

Here's a look at this most reliable of all seasonal trends -- and how I'd play it this year.

Everyone knows that September is historically the worst month of the year for the stock market, with August and October running close behind. In most years since 1950, after a brief rally into the third week in August, stocks have trended lower into September before making a bottom around the 20th of October that sets up the traditional year-end rally.

Doesn't happen every year, of course, and the pattern is subject to annual differences. This year, for example, August turned in an uncharacteristically strong performance with the S&P 500 Index ($INX) up 2% for the month.

Remarkably reliable

But this seasonal trend is remarkably reliable even with these annual variations. In 2005, for example, even though the S&P 500 rallied -- against the historical trend -- from Aug. 15 to Sept. 12, the larger seasonal pattern remained intact. From the close on Aug. 15 to the October low on the 27th, the S&P 500 fell by 4%, despite the unusual early September rally. Stocks then proceeded to climb by 10% from that low through Jan. 11.

The trend is so reliable because:

• There's a structural reason for stocks to slump at this time of the year.

• Stock-market trends that work year after year become self-fulfilling prophecies.

The structural factors that underlie the August-through-October slump begin with the low trading volumes that characterize the stock market in the summer months. Many Wall Street trading desks operate on summer hours. Low trading volumes beget lower trading volumes, since it's harder to get a good price for either a buy or sell when fewer shares are changing hands.

Professional traders who make their money on moves of a point or less don't like trading in markets that force them to give up a hefty part of their profits. Nothing rankles quite so much, no matter what side of the trade you're on, as being told that you can buy at $25, but if you want to sell, the price is only $24.50.

And, professional or not, nobody likes to watch the way rallies tend to peter out, turning the big gains of a morning into a few meager points by afternoon.

As volume picks up after Labor Day, stocks face another structural factor pushing prices generally lower.

Mutual funds operate on a fiscal year that closes on Oct. 31. They have to square their year for tax purposes by that date. So fund managers who want to keep the taxes paid by their fund as low as possible look to sell losing stocks that will offset the gains they've booked over the course of the fiscal year. This tax selling is a constant tide in the stock market in September and October that makes it hard for prices to flow upward.

It's the cessation of that tax selling after Oct. 31 that sets the stage for the traditional end-of-year rally as managers with cash on their hands after those tax-related sales put the money to work again.

Great expectations

Psychologically, the very historical reliability of this seasonal trend becomes a reason for the persistence of the trend. If you know that stock prices will be lower in October than they are at the end of August or the beginning of September, you wait to buy in October. Which, of course, means that stock prices drift down in August and September and are indeed lower in October. And, if you know that the October low marks the low for the season, you start to buy around that low, give or take a few days. And since everyone knows this low is the time to buy, in most years that buying is enough to make stocks move up for the classic fourth-quarter rally.

So how do you play this seasonal pattern in your portfolio?

1. Recognize what won't work. Picking up the gains from this seasonal trend is swell, but you don't want to radically reconfigure your portfolio, gutting positions built up over the long term, because trading costs and taxes will eat up a good portion of the seasonal gains. The 10% gain that buying at the bottom and selling at the top would have garnered you in 2005 is a misleading number that assumes an investor was able to buy at the exact low and sell at the exact high -- and was willing to bet his or her entire portfolio on that predicted timing. Miss either and the returns are significantly lower. Let's say that you bought on Oct. 23, a Monday, instead of Oct. 27, a Thursday, and sold on Jan. 4 instead of Jan. 11. Your return then falls to 6% from the theoretical 10% earned by the investor with perfect timing. Still, 6% in two or three months isn't something I'd pass up.

2. Raise cash by being on opportunistic seller. Think not just about the goal of making 6% or so (hey, it's a goal, not a guarantee) in the end-of-the-year rally, but about what stocks you'd like to hold in your portfolio at the beginning of 2007 -- and for the rest of the year. Sell the stocks that don't fit that strategic vision in any short rallies and then hold onto the cash so that you've got the resources to buy in October. I've been doing that with Jubak's Picks in the short rallies of summer. I've cut back on energy-service stocks, a couple of manufacturing stocks exposed to weakness in the auto and home-building markets, and sold a real-estate play or two.

I've also dumped some stocks that have turned out to be mistakes and taken profits on others where the returns look less promising ahead. After the sale of Plum Creek Timber (PCL, news, msgs) on Aug. 15, the Jubak's Picks portfolio is approximately 22% in cash.

3. Manage your fears. Don't let the reputation of the "worst" months panic you into abandoning long-term holdings or strategies. I'm all in favor of picking up a few percentage points here and a few there from seasonal and other trends, but the biggest gains for most investors come with buying and then letting time work in their favor. Keep some perspective here: The bad months show average losses of 1% to 2% over the last 50 years. That puts them in the "losers" camp because in most months the stock market has been up on average. But it hardly adds up to carnage.

4. Figure out what you want to buy. Take advantage of the down time while you wait for a chance to buy in October to figure out what you want. What sectors will do well, in your opinion, in an end-of-the-year rally and in 2007? (I'm betting on financials and anything that promises safety, such as utility-related stocks. Think Itron (ITRI, news, msgs) or General Cable (BGC, news, msgs), for example, or big diversified multinational industrial or consumer companies like General Electric (GE, news, msgs), PepsiCo (PEP, news, msgs) and Nestle (NSRGY, news, msgs).) Technology stocks traditionally do well in the end-of-the-year rally. Do you think they will again this year? By rallying in mid-August the sector has shown that it can still make a move to the upside. If so, which ones do you want to buy for the short-term? And how do you feel about owning them in 2007? If the economy weakens as many on Wall Street expect, that won't be good for techs.

5. Anticipate the abnormal. Give some thought to what news events might disrupt the normal seasonal pattern -- and what you will want to do about them. For example, after an initial prediction that said the hurricane season of 2006 would rival that of 2005, the actual storm activity has lagged behind forecast. What would you like to own if the initial prediction turns out to be accurate? (Natural gas stocks come to mind.) How about if the storm season is a bust? (Insurance stocks, anyone?) And try to think one step ahead: If hurricanes don't shut down much natural gas production this year, and plunging prices take down the natural gas stocks, would you like to be a buyer in October? At this point, I think the answer is yes -- depending on how low they go.

That's a lot of work to put into figuring out how to profit from just one seasonal trend. But I think the August-to-January move is big enough and reliable enough to make it worthwhile. Standard & Poor's Sam Stovall, who has sliced and diced more seasonal and sector patterns than anybody I know, figures that from 1945 to 2006 the returns from being invested in the S&P 500 stocks came to an annualized 7.1% November to April. For May to October, the return is just 1.5%. I'd work hard for that kind of spread.

After all, Christmas comes but once a year.

Editor's Note: A new Jubak's Journal is posted every Tuesday and Friday. Please note that Jubak's Picks recommendations are for a 12- to 18-month time horizon. For suggestions to help navigate the treacherous interest-rate environment, see Jim's new portfolio, Dividend stocks for income investors. For picks with a truly long-term perspective, see Jubak's 50 best stocks in the world or Future Fantastic 50 Portfolio.

E-mail Jim Jubak at jjmail@microsoft.com.

At the time of publication, Jim Jubak owned or controlled shares of the following equities mentioned in this column: General Cable and PepsiCo. He does not own short positions in any stock mentioned in this column.