| |
 September 2010 Diversification's Holy Grail You Won't Find It Where You've Been Looking
But the one thing everyone does agree on is the benefit of diversification. From the saver splitting CDs between banks to the hedge fund manager mixing currency futures and interest rate derivatives with plain vanilla stocks and bonds, everyone is seeking the most effectively diversified portfolio.
The basic concept is simple: Differing asset classes have different risk and return profiles. Some do well in markets when other fare poorly, so combining them can not only smooth out long-term returns, it can also control volatility or risk. Classes that behave differently -- or in statistical terms those that are not highly correlated -- make the best combinations. Back when Dr. Markowitz' work was first published, there were fewer diversification opportunities. Now that everyone has become a believer, the financial market is rushing to bring out new and different products with the promise of additional diversification for the traditional portfolio. In just one week this past July, exchange traded funds (ETFs) were introduced offering exposure to global equities, the VIX (volatility index), target maturity corporate bonds, international equity sectors, and bull and bear funds for retail and natural gas. That's certainly a far cry from 15 years ago when a seasoned analyst was quoted saying, "You only need U.S. equities and cash to create a well-diversified portfolio." So how are all the new diversification options working? One might think with everyone in hot pursuit of diversification and so many new ways to achieve it, today's portfolios would be the most efficient ever. Then again, virtually everyone suffered in 2008 when it seemed there was no place to hide. With all this new fangled diversification, what happened?
Spicing Up Plain Vanilla Correlations run from +1 (perfect positive correlation) to -1 (perfect negative correlation). A correlation of 0 indicates there is no relation either positive or negative between the two series. Again, in constructing highly diversified portfolios, investors seek to add assets that have low or better yet, negative correlations with the existing portfolio. With this as a background, it's possible to look back over the past to see how well various asset classes have served as diversifiers. Of course the first thing to determine is what they're diversifying. In and of itself, an asset can't be a diversifier, it can only be so in regard to something else, generally a pre-existing portfolio.
So consider the portfolio illustrated in Chart 1. It was built based on the Morningstar categories of mutual funds to represent a plain vanilla stocks, bonds, cash portfolio mix. In this case, it's 30% Large Cap Blend, 20% Small Cap Blend, 10% Foreign Large Cap Equity, and 40% U.S. Intermediate Government Bonds -- the old 60/40 stock/bond mix. This is the type of portfolio many individual investors already have and are encouraged to diversify. Now to find some potential diversifiers. Despite all the esoteric ETFs and alternative investments hitting the market today, it's not as easy as you might think to find some with a track record of five years or more. Again looking at the Morningstar universe, these are the available candidates:
No, these aren't as sexy as bets on the VIX index or inverse funds on natural gas, but they are traditional diversifiers. Bonds have historically moved inversely with stocks. Foreign equities either from emerging markets or world allocations have traditionally been promoted for their diversifying characteristics. Up until the housing bubble popped, real estate was viewed as an asset class all to itself, not highly correlated with either stocks or bonds. Currencies are tied to the relative fate of national economies, not domestic stocks or bonds.
Perhaps more importantly, these were some of the few alternatives with decent track records extending back into the 1990s. This was needed because we wanted to see what happened to the rolling correlations over five-year rolling periods. The actual data set ran from December 1994 through June 2010. Each rolling period contains 60 months (five years). To roll forward, month 1 is dropped and month 61 is added. The actual comparison was based on the correlations between each of the diversification candidates and the base portfolio. In this case, the lower the correlations, the better the diversification. The results are shown on Chart 2.
Three Surprises First, over the past fifteen years, virtually all correlations have converged. That's not surprising given the way technology has vastly improved the spread of information. Bond prices which used to be only accessible from a call to a institutional broker are now easily obtained on the internet. Foreign equity exchanges are now as technologically advanced and efficient as their domestic counterparts. News from the U.S. is reflected in Asian share prices almost immediately. A smaller world has led to higher correlation. But what's not so expected is what happened in 2008. Although correlations had been drifting up over time, they all converged in 2008 -- precisely when investors needed diversification the most. It didn't matter what the series was, all correlations increased when the credit crisis shook the financial markets and virtually everyone suffered. That's not how diversification is supposed to work. In fact, it's the worst of all possible worlds. Just consider, when the base portfolio is performing well, other assets with low correlations will lag behind. If they're included in the portfolio, they'll actually diminish return. You want this behavior when the base portfolio is struggling because the diversifying classes will improve results. But that benefit didn't occur in 2008. Instead, correlations all converged taking all classes down together. Dr. Markowitz didn't predict this. The second surprise comes after 2008. After the initial shock of the credit crisis, equities staged a strong recovery in 2009, but correlations remained elevated. Although they declined from their 2008 peaks, they remained well above their pre-crisis levels. What's up with that? Essentially that suggests asset classes that were previously effective diversifiers now were much less so. As a result, previously well-diversified portfolios were now less so, too. So it's no wonder investors are seeking the newest and greatest ETF; they need new and better forms of diversification. The old ones are becoming less and less effective. Or are they? Take another look at Chart 2. One series has consistently maintained a low correlation with the base portfolio, the 30-day Treasury Bill also known as cash. For a good part of the period, it actually had a negative correlation and even after spiking to slightly over 30% in 2008, it quickly fell back to 5.5%. It's ironic that investors are willing to incur the mutual fund universe's highest expense ratios for esoteric investments with little or no track records when the best diversifier is right under their noses -- where it's always been. Sometimes the simplest (and least expensive) alternatives are the best -- and this is a prime example. Perhaps that seasoned analyst was right. Search this site! Just enter you key word or words:
Get current quotes or follow your own custom portfolio,
courtesy of E-Line Financials:
|
||||||||||||||||||||||
T'S BEEN OVER FIFTY YEARS since Harry Markowitz showed the world that efficient combinations of risky assets could actually help control overall portfolio risk and increase expected long-term return. Ever since, investors have struggled to find the best combination that would produce outstanding returns while minimizing risk on the downside. So far, no one has found the silver bullet.
