Nick Colas: Correlations Ticking Back Up

The idea behind tracking correlations between stocks is to get a sense of how much longevity a particular market trend might have. It is one data point, to be sure, but a good one.

Correlations had been dropping steadily since the Eurosave as the market had moved higher. We saw all kinds of healthy rotation in the tape and non-lockstep action between the various sectors.

But now, Nicholas Colas (ConvergEx Group chief market strategist) is seeing correlations once again ticking higher. Some data from his report this morning:

Every month, we run an analysis of the correlation shown by various asset classes relative to the S&P 500.  Here’s a summary of what we found this time around:

·         The ten industry sectors of the S&P 500 currently trade with an above-average amount of correlation to the index itself as compared to the last +3 years.  As the data table after this note shows, the average correlation for industries such as Financials, Technology, Materials and Health Care is 86.5% since October 2009.  Last month the average was 88.5%.  That’s still within the “One sigma” standard deviation of 6.2%, but distinctly above the norm.

 

·         Most sectors show a higher correlation to the price action of the market index than in October 2012.  Only Technology, Financials and Materials show lower correlations over the last 3 months.

 

·         This is surprising, since over the last four years a rising market usually comes with lower industry correlations.  You’ve no doubt heard the saying that “The market takes the stairs up, but the elevator down.” Well, that elevator is also very crowded, since selloffs have been “Macro” in origin – think U.S. fiscal policy debates or European monetary experiments – and tend to push all risk assets around pretty much in unison.

 

·         On its own, this observation would be a clear negative for stocks.  A “Healthy” stock market shows lower correlations among industry groups, since their fundamentals are different.  The old pre=Crisis benchmark for industry correlations was about 50%.  We are still in the high 80s, and climbing.

 

·         You do, however, have to put the move over the past month – both asset price appreciation and correlations – in a greater context.  Yes, U.S. stocks over the past month are up about 3.6%, or 51 points.  But that aggregate move essentially came in one burst on January 2^nd, most of it before U.S. stocks even opened.  The period from December 14^th to the end of the year was a 40 point churn, first higher then lower.  And since the January 3^rd open, the S&P has been in a 15 point band.

 

·         Many market observers have used the low levels of the CBOE VIX Index to posit that perhaps the ‘Fear Index’ is broken, or that the “Risks” are more than a month away (the VIX’s time horizon), or that things are better than they appear given the headlines about Federal Debt Limits as so forth.  Our own work on the relationship between industry correlations and the CBOE VIX Index shows that the rock –bottom “Fear Index” is in fact out of step with reality.  Over the past few years, the VIX occasionally lingers at strangely low levels, only to come back with a vengeance.  Higher correlations are a good “Tell” of such periods of amplified volatility, and we certainly have those right now.

 

·         We would also note that virtually every other class we track for this monthly correlation study is acting “Normally.”  Gold and silver correlations are at 25-30%, spot on the long run averages.  High yield and investment grade bonds run 81% and (29)% correlations to U.S. stocks, respectively.  Again, a quick look at the time series charts shows these are very much on top of long run averages.  OK, the Euro is acting a bit oddball-ish, with a record low 9% correlation to U.S. stocks.  We’ll see if that lasts longer than  a month before considering what that might mean.

Colas concludes that higher correlations combined with the ultra-low Vix regime put a continuation of the uptrend for stocks in doubt for the short-term.

Bears watching.

Source:

ConvergEx