At the beginning of the year, we took a look at a survey by the American Association of Individual Investors (AAII) which showed that for the first time in that poll's 20-year history, its members put more money into cash than stocks.  There were two other months that came close, and both were excellent long-term buy signals for the stock market.

The latest survey showed that these individual investors reduced their cash holdings by a relatively large 12% last month.  But it wasn't just to buy stocks - they put half of that money into the bond market.

This 22% allocation to the bond market, in spite of historically low yields, is their highest amount since February 1994.  It surpasses the "flight to safety" jump we saw in late 2002 and early 2003, and it's also tied for the largest one-month increase in allocations to the bond market.

Bond prices have been on a steady uptrend over the past 20 years, not undergoing the same kind of sustained swings that stocks have.  In terms of timing the market, changes in AAII member bond allocations have worked OK, but obviously not quite as well on the sell side, so while I think this jump is a warning sign for bond prices (even after their recent drop), I wouldn't count on an end to the bond bull market just because of this.

I've been seeing a lot of excitement over the uptick in the Baltic Dry Index (BDI), which has kind of become the TED Spread of 2009.  The TED Spread was something relatively few had heard of, and even fewer followed regularly, before it became headline fodder during the crisis last fall.

Now the BDI is getting the attention, partly because of its dramatic fall (!) over the past year, and partly because it has rallied nearly 100% since November.  Commentators are beginning to breathlessly report on its daily movements, even though it's an esoteric indicator based on international shipping rates with relatively few inputs and which is subject to manipulation and country-specific structural weaknesses.

Anyway, what we want to know is...does it lead the stock market?

It's hard to see from the chart above, but there is relatively little correlation between the BDI and the S&P 500.  The BDI suffered huge declines during 1995-1996, for example, while stocks rose rapidly.  The same goes for early 2004 and 2005.

Even worse, the BDI was rising right up through 1987, 2000 and 2008, giving us absolutely no advance warning of the trouble that was about to befall the stock market.

The chart below shows the correlation between monthly changes in the BDI and future 3-month changes in the S&P 500.  I'm being generous here, as this is the largest correlation I could find among various changes in the two indices.

The black line that runs through the blue dots is sloping up and to the right, meaning that there is a positive correlation between the two.  In other words, declines in the BDI have very generally led to declines in the S&P 500 over the next three months, and vice-versa.

But the correlation is weak, and it doesn't necessarily imply that changes in the BDI cause changes in the S&P 500.  Also, positive movements in the BDI have not prevented some large future losses in the S&P, as we noted above.

I doubt the BDI factors into the analysis of too many Wall Street analysts, but I was struck recently by just how few of those analysts have "buy" ratings on the stocks they cover.  According to Bloomberg, only 34% of current Wall Street ratings are considered "buys", the lowest since at least 1997.  More than anything, analysts are neutral (60% of the total).

That's quite a change.  In May 2000, they issued 87,000 buy recommendations versus 1,600 sells.  In April 2003 they issued 46,500 buys versus 10,200 sells, still four-to-one in favor of buys but seemingly much more reasonable than at the peak of the bubble.

The chart below shows the percentage of the total that buys, sells and holds make up since 1997.

It had all the makings of a good contrary indicator, until Congress got involved in 2002 in the wake of Enron et al.  Since then, analysts have been much less inclined to stick their neck out either way (to be fair, part of the reason is reduced access to corporate management).  Currently, 60% of all recommendations are Holds, a trend that has set a record every month since September 2006.

Because of changed regulations, I don't think we can consider the current apathy among analysts to be a buy signal.  Generally, I do think the data is more of a contrary indicator than anything, and would be looking for the next jump in "sell" ratings to reflect more of a buy signal than what we're seeing now.

 Disclaimer  |  Privacy Policy