I read an article earlier this week by Alex Rosenberg, "Sector Correlations Falling, Which is a Positive for Traders", that described the recent rapid decline in correlations between sectors and the overall market. Apparently correlations have now come down dramatically from the extreme high correlation a few months ago after the market shock from the swoon in February. The article was saying that this was a good thing because now one can go back to choosing areas of the market based on their own factors and not just based on some big news event. Thus, sector and stock-picking may enable higher returns. The article came with a nice seven year chart of correlation percentages.
As an active investor, I appreciate that lower correlations enables both the opportunity for increased beta and also the benefits of diversification. But, I was wondering whether there is a relationship between movement in correlations of sectors to the market and the market itself that might be useful after seeing the pop in correlations when everyone panicked in February and now the recent move downward as we get farther away from the panic. Is this typical?
Well, here it is - the seven year correlation chart from the article aligned with one for the S&P 500. The answer (drum roll, please) is that yes, correlations seem to relate directly to the market. Lows in the correlation percentage are associated with market tops, and peaks in correlation percentage are associated with a short time period after a meaningful market decline.
Immediately after a big market drop, correlations skyrocket with peak correlations being just a bit after the bottom. This is as you would expect; trading reverts to news-based and trades are made based on guesses in macro factors, not companies or sectors as us human traders try to manage our panic. It is well known that in times of fear our biological processes narrow our focus away from the periphery toward the perceived source of the fear.
But here is the interesting thing - it also appears that when the market sectors are least correlated, the market is at a top immediately before the market drops. This would imply the opposite behavior, of complacency or lack of focus as markets are reaching highs.
So, while the author of the article that I read seemed to be signaling the "all clear" to becoming an active investor again now that correlations have dropped, I can't help but notice that the correlations are at the lowest point in seven years, and that at previous times when correlations declined rapidly as they recently done, a market top occurred followed by a meaningful decline. So, correlations are very low, and at previous times this preceded a drop in the market itself. This is information worthy of raising the Yellow Caution Flag.
But note, none of the market drops in the past seven years are "the big one" as in 2000 or 2008; they have been the normal 10% to 15% variety that we see typically about once a year. So even the current very low correlation reading does not necessarily signal a major market top. A "normal" market drop at this time fits with my trend cycle / small cycle / mini cycle work that would expect a small drop between now and the end of June and a more meaningful drop later this summer.
Disclosure: I/we have no positions in any stocks mentioned, and no plans to initiate any positions within the next 72 hours.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.