In the wake of the financial crisis five years ago, the diversification models practiced and preached by the financial industry have gotten a bad rap.  For decades the conventional wisdom was that that the prudent, intelligent investor should diversify their investment portfolio in order to reduce risks.  And yet when the storm hit in 2007-08, many investors experienced losses much more extreme than anticipated.   The fundamental question being asked is: were the models themselves wrong or were they applied poorly?

I believe the shortcomings were in both theory and practice.  Over the last five years there have been many academic papers written and theories developed with ever-more sophisticated mathematical models addressing tail risk; the risk of rare-but-extreme events.  But just as important is how diversification models get implemented. 
 
Key to the whole idea of diversification is a portfolio should be constructed of asset classes that behave differently at different points in time and with investments that are fundamentally dissimilar.  It was Harry Markowitz’s idea of using uncorrelated assets to reduce overall portfolio volatility that kicked off the whole idea of Modern Portfolio Theory.  But lost somewhere along the way was the flip-side of the idea: if the investments are highly correlated, risk won’t be reduced at all.

Prior to the credit crisis the way many investors achieved “diversification” was to take a single market and divide the pie into ever-smaller slices.  Small caps were separated from large caps, then mid caps were carved out from the two.  The inelegantly named “SMid cap” was added to the mix and microcap, representing but a sliver of the market, became its own asset class.  The lexicon was expanded to include terms like absolute value, relative value, core, growth-at-a-reasonable-price, and momentum growth.  Investors “diversified” by making sure they had at least one manager covering each of these styles.  But the problem was that all of these so-called asset classes or styles were being cut from the same piece of cloth.

No one seemed to mind as the broad market equity market posted a cumulative gain of 2,644 percent from 1980 to 2006.  Sure, some of the aforementioned investment styles would periodically over- or under-perform each other on a relative basis, but as long as the average annual return for the broad market was 13.1 percent it could be claimed that diversification was working.  Of course during the credit crisis of 2007-2008 some investors experienced the downside to having highly correlated investments and saw their portfolios decimated.
 
Today investors are returning to the original intent of diversification- owning investments that truly behave differently.  This can be achieved from the top-down or the bottom-up.

By top-down I mean that investors are open to investing in entirely new asset classes once considered too exotic.  Commodities, precious metals, emerging market debt, hedge funds, and currency strategies are all part of the portfolio conversation these days.  For those with the means and the stomach, asset classes like private equity, farmland, timberland, infrastructure projects, water rights, and carbon rights are potential investments.
 
Of course such investments are not for everyone and require a lot more research and an understanding of the unique risks inherent to each asset class.  Liquidity risk or political risk might be much more relevant than volatility risk for some of these asset classes.  Information is much harder to access.  Traditional valuation models might not apply.  But then again that’s the whole point of incorporating these new asset classes- yes, they have their own risks, but their risks are idiosyncratic and not systematic.  This is how you achieve diversification.

Easing the implementation of this has been the explosive growth of the ETF market, where just about every conceivable strategy or asset class is accessible in the form of an exchange-traded fund.

Alternatively, one can diversify from the bottom-up.  By this I mean investors are selecting money managers which a much broader mandate than the style-pure, low-tracking error strategies that were en vogue during the 1990’s.  These days investors care much less about adding an incremental 1 percent -2 percent over the benchmark if the benchmark is prone to dropping by 50 percent.  Many investors are much more comfortable with money managers with high active share, own concentrated portfolios, rotate between sectors, hold high cash reserves, or market-time between asset classes.  The stigma from such highly active strategies has lessened, just as long as the money manager has a proven track record successfully pursuing such a strategy.

Some investors are splitting the difference by implementing a core-satellite approach.  The core of their portfolio is represented by broad, passively-managed beta products that can be owned at very low cost.  The “satellites” around the core are a collection of these go-anywhere, do-anything managers.  The hope is that by owning a collection of these broad-mandate managers the blow-up risk of any one manager underperforming is diversified away.

That said, if an investor believes diversification remains a worthy goal, they must realize portions of their portfolio will inevitably underperform other portions.  During the period of 1980 to 1999 when the broad U.S. stock market retuned a cumulative 2,269 percent, commodities returned only 303 percent.  It would have taken a lot of will power to hold on to the diversification benefits of commodities during that stretch when the equity markets were skyrocketing.

It’s a simple truth, but if everything in your portfolio is going up at the same time, you’re not diversified.

Marc Odo is Director of Applied Research at Informa Investment Solutions, a leading provider of software and strategic solutions to the financial services marketplace.