Bull and Bear Market Lengths

The financial crisis seemed like it wouldn't end, so it's pretty hard to believe that the bull market that started back in March of 2009 has already been going on for longer than the ugly bear market that preceded it (10/07-03/09). 

Below we highlight the lengths of every bull (green) and bear (red) market for the S&P 500 going back to 1928.  The dates represent the start of each bull and bear.  As shown, the average bull market (915 days) has historically lasted about three times as long as the average bear market (310 days). 

Prior to 1940, we saw a number of short market cycles, but since then they've lasted much longer.  In fact, since 1940, the average bull market has lasted more than 1,600 days, which is about 2.5 times as long as the one we're currently in. 

 

Comments

[kobus]

Hi there can you give your parameters for a "bull" and "bear" market please

Thanks

Kobus

[Bespoke]

Sure Kobus. Bull and bear markets are defined as 20%+ rallies from a trough and 20%+ declines from a peak.

[pflynnharv]

Why not use the median instead of the average unit. It is a more appropriate assessment of the length of bull and bear markets. As you know outliers influence the average measure; so, the comparison of the length of bear or bull markets losses some of its relevance. Moreover, the relative length of bull to bear markets will likely be significantly less than three times as suggested by the average measure. It would be interesting to see the level of operating margins peaks and troughs overlaid with their respective bull and bear markets. It should show a representation of the rate of change in margins prior to their peaks or troughs in the content of the bull and bear markets. The reversion to the mean is relevant when reviewing this analysis.

[kindleburger]

Foreshortening the bull market line from '87 to '00 hurts the graphical representation because the current market bust is a function of that 4,494 day boom. It suggests that the current bust will be of longer duration given the inevitable mean reversion.

[Swag]

Isn't it a bit more common to define a bear as a 15% reduction? Maybe as a nod to Dow theory. It changes your bull market counts and lengths a great deal, despite the small difference. It basically makes both bull and bear moves a great deal more common, and they last much less time: I can't remember what they were exactly, but I think they lasted less than half as long as what you state here. Of course, any definitional change will alter averages and counts, but for me, that difference is quite significant with regards to strategy.

[paulflyharv]

Why not use the median instead of the average unit. It is a more appropriate assessment of the length of bull and bear markets. As you know outliers influence the average measure; so, the comparison of the length of bear or bull markets losses some of its relevance. Moreover, the relative length of bull to bear markets will likely be significantly less than three times as suggested by the average measure. It would be interesting to see the level of operating margins peaks and troughs overlaid with their respective bull and bear markets. It should show a representation of the rate of change in margins prior to their peak or trough in the content of the bull and bear markets. The reversion to the mean is relevant when reviewing this analysis.