# What Declining Stock Correlations Mean for the Market

by Daniel Putnam | May 1, 2013 9:28 am

Correlations among the individual stocks that make up the U.S. stock market can be a useful indicator for investors, but it isn’t an issue that receives a great deal of attention. In all likelihood, this lack of press is probably because the process of calculating correlation represents the “deep end of the pool” in terms of esoteric option math.

But you don’t have to be a mathematician to incorporate the concept of correlations into your toolbox.

### Correlation 101

First, here’s a quick, math-free run-through of what correlation is within this context.

Correlations are measured by the CBOE S&P 500 Correlation Indices, which compare the implied volatility of index options to the implied volatility of the 50 largest individual stocks in the index.

The CBOE website explains that while the IV of an individual stock is determined solely by expectations for that stock’s future volatility, the implied volatility of a stock index is determined by two factors: the total of the implied volatility of its components and the expectations for the degree of future correlation among individual stocks.

The Correlation Indices measure the difference between the two implied volatilities, which carves out the element of index-option pricing tied solely to expected correlations.

A full, detailed (and plain-English) explanation can be viewed here[1].

### How to Use Correlation Data — and Where to Find It

Why would anyone bother with something so obscure? Very simply, because correlation statistics can be used as a rough gauge of market sentiment — and as a companion to the VIX.

When correlations rise, it’s usually because investors are dumping stocks in response to a particular event — say, one of the periodic revivals of the European debt crisis. Typically, such occasions tend to be a buy signal.

On the other end of the spectrum, lower correlations show that investors are less concerned with macro events and are judging stocks on their individual merits. While low correlations aren’t a sell signal on their own — since, like market rallies themselves, they can persist for quite some time — they can provide some insight into the state of investor sentiment.

This is evident by the decline in correlations that has occurred during the past year. As the concerns about Europe slipped into the rear-view mirror and investors have grown more confident in the growth outlook and the odds that central banks will provide endless stimulus, correlations have dropped from the above-average levels at which they have spent most of the post-financial crisis era.

Currently, the chart shows correlations sitting close to their recent lows. This isn’t surprising given that the market itself is at an all-time high, but with a wide dispersion in returns among the defensive and cyclical sectors.

Keep in mind, however, that this index will shoot right back up the next time investors go into “risk-off” mode.

[2]

The low level of the index right now indicates that it should be easier to add value through individual stock selection. When correlations are high, individual stocks are much more likely to track the direction of the broader market, meaning that a correct fundamental view on a particular stock isn’t necessarily going to be reflected in that stock’s actual performance.

This might be one of the reasons why Standard & Poor’s, in its 2012 S&P Indices Versus Active Funds (SPIVA) Scorecard[3], calculates that a full 86.5% of large-cap funds underperformed the S&P 500 Index in the three-year period from 2010-12 — a time characterized by high correlations.

If active stock managers are ever going to earn their paychecks, this year-to-date period — when they are unable to use high correlations as an excuse for poor relative performance — should be their time to shine.

### The Bottom Line

Watching the Correlation Indices isn’t going to make you rich, and as yet there’s no ETF to track it as there is with the VIX. Still, it represents another way to monitor investor sentiment and assess the performance of active managers.

Endnotes:
1. here: http://www.cboe.com/micro/impliedcorrelation/