Mebane Faber portfolio comparison

[kevink]

Succinct and interesting overview of 8 sophisticated portfolios. Especially interesting (to me, anyway) are the maximum drawdowns, since actual risk tolerance often proves to be a whole lot less than hypothetical when markets crash, often turning buy-and-hold into panic-and-sell at the worst possible time.

http://http://www.mebanefaber.com/2013/07/31/asset-allocation-strategies-2/

 

[pb4uski]

Link doesn't work but this one does.

Asset Allocation Strategies – Mebane Faber Research – Stock Market and Investing Blog

 

[haha]

This little study, as well as almost all other public investment strategies are backward looking. If one could do it again over the next 40 years there would likely be no correlation with the results reported here- except that there may still be no clear winner, depending on how one picks winners.

IMO this will always be so, since people love numbers and hate the unknown, and hate to take what seem to be risks. Most of us will have about as much luck with these backward looking strategies as the French had with the Maginot Line, or in an earlier war with the cavalry charge into German machine guns at the Somme.

People look for comfort, which at best is probably minimally correlated with effectiveness.

Ha

 

[kevink]

This little study, as well as almost all other public investment strategies are backward looking. If one could do it again over the next 40 years there would likely be no correlation with the results reported here- except that there may still be no clear winner, depending on how one picks winners.

IMO this will always be so, since people love numbers and hate the unknown, and hate to take what seem to be risks. Most of us will have about as much luck with these backward looking strategies as the French had with the Maginot Line, or in an earlier war with the cavalry charge into German machine guns at the Somme.

People look for comfort, which at best is probably minimally correlated with effectiveness.

Ha

Pretty much every investment approach discussed and recommended on this board and most other places is based on backtesting using historical performance. FireCalc is also 100% based on this. Actually the ONLY portfolio in the Mebane Faber article that isn't based on backtesting is Harry Browne's Permenent Portfolio, the weirdest (for most) allocation of the 8, though in fairness I would guess most of the relatively few who've implemented it have done so in part based on its historical performance since the '70's and in particular the relatively low worst-case drawdown. "Past performance is no guarantee of future results," but clearly it's human nature to use - and rely on- the past.

 

[ejman]

Pretty much every investment approach discussed and recommended on this board and most other places is based on backtesting using historical performance. FireCalc is also 100% based on this. Actually the ONLY portfolio in the Mebane Faber article that isn't based on backtesting is Harry Browne's Permenent Portfolio, the weirdest (for most) allocation of the 8, though in fairness I would guess most of the relatively few who've implemented it have done so in part based on its historical performance since the '70's and in particular the relatively low worst-case drawdown. "Past performance is no guarantee of future results," but clearly it's human nature to use - and rely on- the past.

If one can't rely on the past - what's left? Well, there is the present - a fleeting phantom. And beyond that? an unknowable future. I guess I'll stick to what has worked reasonably well in the past until someone comes up with a reliable way of getting tomorrows news today. To follow up on the thought, it would be against human nature not to use and rely on the past.

 

[LOL!]

I think the point of the article was that everything worked pretty much about the same. All the discussion and gnashing of teeth to get the best portfolio just didn't matter in the end because they were all the best portfolio.

And that makes sense. No one would tout a portfolio that would obviously quickly fail. For example, we don't see books about the "lottery ticket" portfolio: "Just take your entire nestegg the day you retire and buy powerball tickets."

 

[kevink]

Actually the key take-away for me isn't the similar long-term performance of the portfolios, though that's interesting enough. The key issue as I mentioned when I posted it is the difference in the worst-case draw downs, because very few investors will in fact stay the course in the face of 30-40%+ drops. For early retirees this is an especially important issue, IMHO, since not only to you need to make your money last over an exceptionally long period but your ability to recover from a catastrophic loss, either through paid employment or the passage of time, does nothing but diminish.

As for ejman's comments, rather than relying on the past one could look at it carefully and then rely more on truly uncorrelated asset classes. That's what the Permanent Portfolio does, and that's also what folks who keep a significant chunk of their investable assets out of the markets do (e.g. by owning income properties, keeping a chunk in a privately held business, etc.). That's also the mindset behind annuity investments, though clearly there counterparty risks and other complexities murky the waters.