Courtesy of Doug Short.
It’s mid-week in the latter part of August. Economists have been ratcheting down their estimates for Q3 GDP in wake of several signs that the recovery is stalling. Volatility in the market has increased substantially over the past few weeks, and my favorite US market index, the S&P 500, has been trading in a range approximately 12 to 18% off the interim high of late April.
What are the pundits now focusing on? Chairman Bernanke Friday speech from Jackson Hole, Wyoming, and President Obama’s post-Labor Day address from Detroit.
Many investors are hoping for another Jackson Hole hint of more quantitative easing from the Fed — essentially a repeat of last year’s goose to the market. Others are speculating on what sort of job-creation stimulus the President will unveil in his September 5th speech in Detroit for the annual celebration of workers.
Let’s step for a moment into the realm of market trivia and add another item to the watch list — the October Effect in the stock market. Last week I offered a retrospective on 4% plus declines in the market since 2000. In my closing sentence I made a remark about October being “the month that has most often hosted market bottoms.” A reader emailed me for some supporting details. So, let’s check out the chart.
Since 1950, the S&P 500 has had 10 major tops and bottoms. The tops have been spread across a range of months, with August and November each represented twice. As for the bottoms, October has the dubious distinction of hosting five of the ten. On a more positive note (sort a of whimsical Market Almanac Musing and not to be taken too seriously), Haloween seems to be a good bet for reentry after selling in May, otherwise known as the Halloween Indicator.
Note: The 1950 start date uses spiced data from the S&P 90 for the index prior to the launch of the 500 in March 2009. This is the widely used data series made available by Yahoo Finance. The troughs are based on daily closes with a -20% or more qualifier, with one concession — the 19.9% decline in 1990, which would have been greater if we were using intraday highs and lows.
