• Markets
• Economy
• Federal Reserve
• Housing
• Energy
• Digital Media

Sentiment? Depends on who you ask

Our last discussion of sentiment (see Signs of a Market Bottom?) generated a heated debate about whether everyone was too bullish or too bearish.

The short answer is: It depends upon who you ask.

For example, the WSJ's Marketbeat just reported:

"Optimism abounds among portfolio managers heading into 2007, according to Merrill Lynch’s most recent survey of global fund managers. While they believe global growth and corporate profits will weaken in 2007, the nightmare scenarios are being discounted, and investors see the so-called Goldilocks scenario coming to fruition as the pace of economic growth declines without a recession. “Compared to three months ago, 8% of our panel no longer think we are in a late-cycle environment, but are instead still in the mid-cycle phase,” they write.

Overall, fund managers see global growth receding in 2007, and for earnings to decline. But they believe corporate balance sheets are in good shape, with a net 55% saying balance sheets are under-leveraged. Oddly, while managers believe, on average, that stocks will be higher 12 months from now and believe equities are undervalued, most investors are taking slightly lower-than-average risks in their portfolio, suggesting a somewhat defensive stance — and their sector allocation proves it. Investors are overweight in banks, pharmaceuticals, energy and insurance. They’re underweight in consumer discretionary stocks and industrials."

Compare that cohort (albeit a professional and important one) with this subgroup: Bloggers: Ticker Sense's 2007 Financial Blogger Outlook shows a very modest forecast:  Financial Bloggers Expect the S&P 500 to Rise 2.39% in 2007.

~~~

When asking (or answering) the question "What is market sentiment like" remember that your answer depends upon who it is you are surveying . . .

Wednesday, December 20, 2006 | 11:35 AM | Permalink | Comments (27) | TrackBack (0)
add to de.li.cious | digg this! | add to technorati | email this post

Comments

An equally key question is "how do you define bullish?"

You, and the majority of commenters here, apparently define "bullish" as "any positive return on equities whatsoever." My evidence for this statement is that the Bloomers article you referenced had two participants citing returns below 3%, and a consensus below 6%, and you several times referenced it as "bullish" or "unanimous bullishness" in paraphrase. Let's call this "definition Barry." [BR: please don't put words in my mouth]

By using "definition Barry," you have been able to put up a wall of angst that "everyone is bullish," which is, by comparison to typical stock market returns, quite nonsensical.

Some of us have a different definition. We would refer to a projection that is significantly above the typical return on equities as bullish, a projection that is significantly below the typical return on equities as bearish, and any projection that is near a typical return as neutral. Let's call this "definition Bill."

We can quibble, and as a matter of fact, have quibbled, almost endlessly over *exactly* what the average annual stock market return is, and what time period it should be measured over, etc. Enough. It suffices to say that a 5.5% return is below any reasonable measurement of typical returns on equities. The 2.4% is, as well. The survey of Business Week participants put in a consensus 8%, which is, depending on time period used, either in or slightly below the typical range.

By "definition Barry," all three, Bloomers, BW, and Bloggers, are "bullish" because they all expect a positive return. By "definition Bill," using the most conservative (read: low) estimates of typical returns, you have two "bearish" and one "neutral" survey. Using a 1980 to present sample for "typical," all three are "bearish."

I would like for you to directly address what you consider "bullish," "bearish," or "neutral," in terms of percentage returns. I ask for percentage returns because it avoids a debate over time frames for measuring "typical" and allows you wide latitude to choose broad ranges. I don't think that a comparison to a crowd is meaningful because (1) it depends on the crowd, (2) trying to measure crowd sentiment on a curve is self-referential; only individual sentiment can be measured on a curve, and (3) the crowd's sentiment changes over time.

Posted by: Bill a.k.a. NO DooDahs | Dec 20, 2006 11:54:14 AM