Looming over the market this week is the shadow of doom. For it was in the third week of July last year, and in 1998, that all three major stock market indices topped out for the summer. And if you want to go back a little further in time, the market hit its summer high this week in 1990 as well.
Leveraging these numbers with all their might, bears have spent much of the past month pumping the myth of the summer slump in stock prices as if it were a market constant -- something that everybody knows always happens. I'm sure you've heard these voices, even if you weren't paying close attention. Among the most prominent recently was fund manager and newsletter publisher Michael Murphy, who told readers at the start of July to sell tech stocks immediately.
A look at the historical record of the past decade, however, presents a different picture: The past two years' summer setbacks were an anomaly -- not the norm. The
has advanced sharply in eight of the past 10 summers.
If there is anything to the myth of a warm-weather wane in prices, it's really more of an August (and sometimes September) phenomenon than a broad summer phenomenon.
Signs of a Climb
The pattern of trading in this year's market bears more resemblance to 1994 than to 1999 and 1998 -- and that year saw the broad market averages climb steadily through the summer months.
These conclusions probably won't find much traction among folks whose record of investment experience starts around the time of the
IPO in spring 1998, but you don't really have to be an old-timer to benefit from a good look at the facts. So here they are:
Since 1985, the third quarter (July, August and September) has on average been the slowest of the year for the
, the Nasdaq and the
Dow Jones Industrials
-- but it's seldom negative. On average, the third quarter has delivered advances of 0.57% on the Nasdaq, 0.1% on the S&P 500 and 0.07% on the Dow.
Since 1985, June and July have resulted in sharp advances on average for all the indices, while August is the only consistently negative month for every index. September is negative on average for the S&P 500 (minus 0.19%) and the Dow (minus 0.41%), but it's modestly positive for the Nasdaq (plus 0.87%).
The Nasdaq's poor record in Augusts is largely colored by two disastrous years: 1998 (when it lost 19.9% in a month that saw the Asian and Russian financial crises climax) and 1990 (when it lost 13% in the month that Iraq invaded Kuwait, precipitating the Gulf War). Smooth out those two outliers, and even August is actually benign for the tech-heavy index. It gained, respectively, 5.4%, 6.0%, 1.9% and 5.6% in the four years beginning with 1993.
Actually, as long as your view of market history isn't clouded by the most easily retrieved memories, it's much easier to make the case for tech-stock rallies in the summer than tech-stock routs. To pound home the point, let's review the dates of the highs and lows of each year in the past decade. I'll assign points to our competing myths (summer slump vs. summer rally).
In 1999, the three major averages peaked on July 16 or 17. The Nasdaq then fell 13% until the second week of August before reversing. It proceeded to rally 63% through the rest of the year. (Score one point for the summer rally myth, because the Nasdaq actually advanced 10% from June 21 to Sept 21.)
In 1998, the three major averages topped out on July 17 or 20. The Nasdaq then fell 30% until the second week of October before reversing and advancing 45% through the rest of the year. (Score one point for the summer slump myth; the Nasdaq fell 8% from June 21 to Sept 21.)
In 1997, the major averages fell sharply in February, March and April, then reversed and climbed without pause until the second week of August. The Dow dropped sharply at that point, but the S&P 500 and Nasdaq merely yawned. The next high for the averages was early October, and of course there was a crash on Oct. 27. (Score one for a summer Nasdaq rally of 16%.)
In 1996, the Nasdaq peaked on June 5, then fell 16% until the third week of July. It then rose by 19% through the end of the year. The other two indices had only a brief pause in July amid a generally steady year. (Score one point for a mild summer Nasdaq rally of 4%.)
In 1995, the indices all rose dramatically from New Year's Day until the second week of September with just a brief hiccup in mid-July. The Nasdaq was up 43% by Labor Day. The markets then leveled off until 1996. (Score one for a summer Nasdaq rally of 15%.)
In 1994 -- the last year that the
repeatedly raised interest rates -- there was a serious decline in the spring that started March 17 and bottomed on April 19 (sound familiar?). A brief recovery ended with another dramatic decline in late June, but that was followed by a July and August in which the major indices rose without stopping for breath. The indices all saw their top for the year in the second week of September. (Score one for a summer Nasdaq rally of 7%.)
In 1993, the indices experienced their obligatory plunge in April, then went straight up until the end of October. (Score one for summer Nasdaq rally of 7%.)
In 1992, the indices went their separate ways, so it's not easy to generalize. The Nasdaq fell steadily from February to late June, then reversed and slowly gathered steam through the summer before really hitting the ramp up in November. (Score one for a summer Nasdaq rally of 7%).
In 1991 -- well, ya shoulda been there. The market bottomed in January when the ground war started in the Persian Gulf, then steadily rose all year with a brief hiccup in July. (Score one for a mild summer rally in the Nasdaq of 9%.)
In 1990, the indices all peaked on July 16 (same week as 1998 and 1999), and then declined briskly until reversing in the third week of October. (Score one for summer Nasdaq decline of 21%.)
I count two points for the summer decline myth vs. eight points for the summer rally myth. So far in the summer of 2000, the Nasdaq is up 4% -- putting the odds in favor of more sunny weather ahead for tech stocks. My guess is that investors can use inevitable squalls as buying opportunities for long-term holds.
Indeed, Rex Takasugi, an investment manager and publisher of the
newsletter in Kent, Wash., says he's looking for new highs in the indices by mid-August -- and believes that anyone looking at 1998 and 1999 for guidance will be caught flat-footed. "You need to understand how differently the market is acting this month relative to the way it was acting a year ago at this time," he said in an interview. "The spring plunge in the Nasdaq was entirely different than what we saw in 1998 and 1999 -- and set us up for a lot more strength this summer." Takasugi specifically likes Nasdaq 100 stocks because of their liquidity, as well as beaten-down small- and medium-sized technology stocks such as
The tendency to ascribe long-term trend status to one's most recent memories is known to psychologists as the "availability heuristic." The phenomenon is deftly described in a
Web page devoted to forms of cognitive thinking in the online courses of Fairmont State College (W. Va.) provost and psychology professor, Frederick G. Fidura. The concept offers a good explanation for why we are often fooled by our faulty memories of market behavior. "Because memory for items generally declines with time, more recent items are recalled more accurately," Fidura notes. "Since recent items are more available, we judge them to more likely than they really are."
At the time of publication, Jon Markman owned or controlled shares in the following equities named in this column or listed in the SuperModels portfolios: BroadVision, Cisco Systems, Digital Lightwave, Emulex, Kopin, Maxygen, Microsoft, Nokia, Nortel Networks, Oracle, Qualcomm, Siebel Systems, SDL, Superconductor Technologies, Veritas Software and Xcelera.com. Under no circumstances does the information in this column represent a recommendation to buy or sell stocks. He welcomes your feedback at
Rowland's Watch Portfolio