CHAPEL HILL, N.C. (MarketWatch) — Do you (or your clients) think you can get out of stocks in advance of the next bear market?
You (or they) had better think again. The odds of sidestepping a bear market are even lower than previously thought.
That’s saying something, since those odds were already known to be dismal. But, poor as those odds are, they still leave a loophole through which die-hard believers can still wriggle. That loophole should be closed.
This loophole exists because, however depressing the historical performance rankings have been, there inevitably are a select few market timers who succeed in beating the market over an entire market cycle. On the assumption that they can repeat their market-beating performance during the next cycle, the argument goes, does it matter that the rest of the market timers are failures?
But is that assumption a good or fair one? How many of the successful market timers during one market cycle are able to anticipate the subsequent bear market?
That’s what I set out to study for this column.
I started by calculating stock market timers’ returns over the full market cycle beginning with the market top in March 2000 and ending at the October 2007 market top — encompassing both the 2000-2002 bear market and the ensuing bull market. I converted every timer’s rank to a percentile, so that the top performer was given 100 and the worst earned a 0. I then repeated this exercise for performance during the 2007-2009 bear market.
Note carefully that I focused on just the market-timing component of the timers’ performance. That’s because I didn’t want these rankings to reflect their abilities (or lack thereof) to pick particular sectors or stocks.
The accompanying chart summarizes what I found. Take, for example, the first column in that chart, which reflects the 10% of market timers with the best performance over the market cycle from March 2000 to October 2007. Their average percentile rank for performance during the 2007-2009 bear market was just 46.4%.
Ouch. If you had picked a market timer at random, your expected percentile rank for performance during that 2007-2009 bear market would have been 50%. So the best performers between 2000 and 2007 performed worse than random in the subsequent bear market.
To drive home this point even more, notice the second column in the chart, which reflects the 90% of market timers that weren’t in the top decile for performance between 2000 and 2007. Their average percentile rank during the 2007-2009 bear market was higher than it was for the top decile.
In other words, as far as sidestepping the 2007-2009 bear market is concerned, the previous cycle’s worst performers did better than the best.
As you can also see from the chart, a similar — though even starker — contrast emerges when we separate the market timers according to whether they beat or lagged behind a buy-and-hold investor over the 2000-2007 market cycle.
Does this mean that no stock market timer will ever be able to beat a buy-and-hold investor over two successive market cycles? Of course not. It’s always possible, even if unlikely.
The point of this column is that even the die-hard believers in market timing need to think twice, nay thrice, about their confidence that they can get out of stocks before the next bear market. Their confidence is truly a triumph of hope over experience.
I often get asked about why I would report these dismal statistics, since my performance tracking business presumably is built on the assumption that it’s important to separate the top performers from the bottom ones. And I admit it would be far better for my business to report that a sure-fire way of beating the market is to follow the past’s top market timers.
But that’s not what my data show.
For more information, including descriptions of the Hulbert Sentiment Indices, go to The Hulbert Financial Digest or email firstname.lastname@example.org.