THE CROWD IS ALWAYS WRONG DURING MARKET EXTREMES :Business Strategy /Financial Advisors

Here, in the summer of 2005, my sense is that many hedge-fund investors
believe that the equity markets have rallied a bridge too far,
are scared, and have done some selling. The complacency everyone
was talking about seems to have been only skin deep.They all recite
the same mantra that almost all of the sentiment indicators show a
dangerously high level of bullishness. Therefore, they are contrary to
the consensus and are bearish. Being a contrarian is very chic. The
only trouble is that now everyone is a contrarian. Even Wall Street
economists are now contrarians. Goldman Sachs recently published a
paper on inflation that began, “Our new forecast for inflation is out
of consensus.” In other words, the writer thought the most important
aspect of his forecast was not his economic analysis but rather that it
was out of consensus.Therefore, instead of being contrarians, perhaps
we should be contracontrarians.

The sentiment indicators are not easy to decipher. Ned Davis of
Ned Davis Research has studied them for years, and he points out
that the crowd is always wrong at market turning points and that they
are always wrong at extremes in sentiment. He says: “If one knew for
certain the peak or trough in sentiment, then one could go contrary
and be correct nearly all the time.” However, because one cannot
know exact extremes for certain except in hindsight, one can be
weeks or months early, which can be very painful. The Ned Davis
Sentiment Composite began showing extreme optimism in January
1999, then really extreme optimism in December, but the bubble
didn’t burst until the late spring 2000.That was an eternity for those
who sold in January 1999 and very painful for people like me who
reduced allocations to technology in December 1999. However, in
1987, the Ned Davis Composite spiked to an all-time high just a
month before the Crash. It’s the same with bottoms. In August 1990
the Composite reached extreme pessimism, but that was 47 days and
12% on the S&P 500 before the low.

Ned’s advice is to use stop losses. Easier said than done. Most stoploss
orders are put in at 10% limits. Admittedly in the bubble run-up,
speculators who went short or sold too early would have been stopped
out or gone back as prices soared, but when would they have known to
sell again? And in the 1990 example, they would have bought in August,
and then, when the market fell more than 10%, they would have had
their stop-loss order activated just as the turn was finally about to occur.
The timing of the execution is crucial. Often when you leave a stoploss
order with a dealer or when it is on the specialist’s books, you get
picked off by the market maker. In other words, you get an execution
just before the asset in question rallies.