essays pieces of the puzzle
Thursday, November 15th, 2012
stocks bonds cash

Back in the day, sell-side firms put out recommendations for the percentages of stocks, bonds, and cash that investors should have in their portfolios.  (Check out this advertisement for a 1992 Paine Webber seminar; note also “The Big Shift to Total Return Investing” report that was available.)

Strategists did this routinely and after a time people started tracking the aggregate recommendations as an indication of where the Street stood (and often using them as a contrary indicator).  In doing research for my recent articles about cash I stumbled upon the Bloomberg data for the chart above.  The lines show “the average recommended allocation for bonds, stocks, and cash for U.S. chief strategists at Wall Street firms.”

It’s fun to go back and see how wrong this has been at turning points, but what are you looking at, really?  These kinds of numbers get reported as if they are really important, often without key details like what “bonds” really means across the strategists surveyed.

You might notice that the most recent readings don’t add up to 100%.  So, I went looking for the answers.  I found that these numbers are based upon four — count ‘em, four — firms, with these allocations (bonds|stocks|cash, according to Bloomberg):

Bank of America:  50|35|3, with 12% in commodities.

HSBC:  21|54|0, with 9% in commodities and 16% in gold.

J.P. Morgan:  60|25|15

UBS:  47.8|40.7|0.2, plus 6% in commodities, 5.8% in real estate, and -0.5% in volatility.

That’s why the chart numbers don’t add to 100%; other things are left out.

Where to start with this?  How about “garbage in, garbage out”?  The construct at one time made some sense, but there aren’t many strategists left and the strategies themselves are (for better or worse) much different than they used to be.  Yet, you still see the numbers referenced.

You’d be surprised how often that’s the case with “indicators” that are cited by market participants and financial journalists.  Someone should check in every once in a while to whether they make sense to use.  (Chart:  Bloomberg terminal.)