The Investors Intelligence survey has reached a bearish skew not seen since the bottom of the financial crisis
The combination of a significant pullback and increased volatility in stocks has made investment advisors extremely bearish, at least according to the most recent sentiment survey by Investors Intelligence (II). The survey is a collection of over 100 investment newsletters, email bulletins, etc. from investment advisors. The editors at Investors Intelligence determine the percentage that are bullish, bearish, or looking for a correction. To get a single number summarizing the amount of optimism/pessimism in the poll, we often take the percentage that are bullish and subtract the percentage of bears. The bears now outnumber the bulls by the largest percentage since March 2009, the month that stocks bottomed after the financial crisis. This week I'll look back to see if we can get an idea about the chances that another major bottom is in place.
Bulls Minus Bears: Here is a chart of the S&P 500 Index (SPX) along with the difference in the bulls and bears in the Investors Intelligence poll. The green horizontal line is where the bears outnumber the bulls by 10%. In this chart going back to 2000, it looks like this signal marks pretty good times to buy.
Many times quantified results do not show what I think I see on a chart, so in the tables below I attempt to quantify exactly how good of an indicator this has been. Specifically, I went back to 1963 -- as much data as we have on the Investors Intelligence poll -- and found the incidents where the bulls minus bears line fell below the negative 10% mark for the first time in three months (the first such signal occurred in 1965). Then I found the returns on the SPX, going forward over different time frames. The second table below shows the typical index results for comparison.
The indicator does show some bullish returns after a signal, compared to the anytime returns -- the average return after a signal beats the typical average return across all time frames. The shortcoming of the indicator is that the increased returns seem to come at the price of increased volatility. The standard deviation over the next year is higher after a signal, as compared to anytime.
I thought I would take a look at it one more way to see if I could find a better signal. Instead of making a signal when the difference crosses the negative 10% mark, I delayed the signal to when the bulls once again overtake the bears. I thought it might catch better signals because it wouldn't signal until the market settled and began moving higher. The table below shows the results. The standard deviation is lower in the first month returns, but not too much better after that. The average return after one of these signals lags in the short term, but then slightly outperforms when you get to six months and longer. It seems holding off for this signal would have caused you to miss some early gains, but you have done just as well in the longer term.
Finally, I decided to look at more recent time frames. If I do the original indicator above (bulls minus bears moves below negative 10%), but only go back to 1990, the indicator becomes a pretty strong buy. Again, you get a lot of volatility (even more than the first table), but the average returns are even better compared to typical returns. At six months you get an average return more than two and a half times more than the typical market (10.8% vs. 4.1%). Overall, this indicator is giving a decent buy signal, but keep in mind it's at a cost of some increased volatility.
