When to Buy? When to Sell?

Will be in Phoenix/Scottsdale the rest of the week.


Ha!  Meb Fabor….I guess that is not as bad as Melanie Faber when I first published the GTAA article. . .


People often ask me if buy and hold is dead (or make comments about my model "disproving" buy and hold).  Look, I have no problem with buying and holding asset classes – being long risk has proven to be a great strategy for a long, long time. 

I see the issue as more of a personality issue.  With a diversified portfolio of world asset classes (or a 60/40 portfolio) I think you need to be able to accept a 50%+ drawdown.  With a single asset class or security you need to be able to accept an 80%+ drawdown.  I think using risk management via a trendfollowing method fits me personally.  While historical drawdowns have been low for the timing model, I can certainly see scenarios where they could be higher.

The tactical model is meant to only be broadly represenative of a risk management strategy.  It is not the Holy Grail by any means.  One of the main difficulties people have is in doing wholesale shifts in asset classes.  ie "I’m in stocks", or, "I’m out of stocks".

A potentially better option for many investors would be to scale in their investments.  So, instead of using just the 10-month SMA for each asset class, divide the capital into two buckets.  Time half with the 3 to 6-month SMA and half with the 9 to 12-month SMA (take your pick, 50 and 200-day SMA could be similar).  That way you avoid all of the personal stress of "missing rallies" and "sitting in bear markets".  And yes, fyi, I expect there to be no difference in the long run to performance either way.  But for a lot of people it might fit their disposition better.

I have received a ton of emails asking if it is possible to combine the rotation and timing models.  More in a follow up post if I can get around to it. . .