“The problem of correlations is growing, and I don’t think it goes away.” -Joseph Mezrich, head of quantitative research at the U.S. brokerage unit of Nomura Holdings.
The quote above was taken from a Bloomberg article by Eric Martin and Michael Tsang in which they point out that the correlations of stocks, commodities and emerging markets are again moving in lockstep. In fact, the correlations between the S&P 500 and the Reuters/Jefferies CRB Index of commodities over the last sixty days is now at the highest level in 5 decades.
It’s always entertaining to see people extrapolate very recent performance, correlations, or standard deviations in to the distant future. That is quite a statement by Mr. Mezrich that high correlations are never going away.
Check out the following chart of the correlations of the S&P 500 and the MSCI EAFE over a 20-year time horizon.
(Click to enlarge)
Contrary to Mr. Mezrich’s opinion, I think we can expect correlations to do what they have always done in the past: change. Any strategy whose success is dependent upon the stability of performance, correlations, and standard deviations is very likely to experience risk and return characteristics wildly different than they imagined. Data is readily available on each of those factors showing their wide variability over time.
In contrast, Tactical Asset Allocation is not dependent on the stability of performance, correlations, and standard deviation. The success of a Tactical Asset Allocation approach is only dependent upon trends. And trends aren’t going away.
