All That Glitters / Is Diversification Dead?

Remember the conversations you had with investment advisers a year ago? "It's impossible to second-guess the market," they'd announce gravely. "What we have to do is build you a personalized portfolio, diversified in different avenues."

They had good reason to tout diversification. Half a century ago, years of academic research and experience seemed to show that the only free lunch in the marketplace was diversification over a great many investment avenues. Mainly, diversification into avenues that had as little as possible to do with one another.

Analyzing the correlation between investments allowed investors to find their ideal levels of risk and return. Even amateur investors learned about "beta," the correlation of a single investment with the entire market, and "alpha," the surplus returns a good investment manager manages to achieve. A stock's beta is a way to compare its market risk with that of other stocks.

The entire industry of investment management was based on alphas and betas, and diversification across sectors and countries, across different types of derivatives and financial assets, and whatnot.

Naturally, it played into the hands of the investment banks. The process looked scientific. It boasted mathematical formulae and terms that were very hard to explain. It required a great deal of buying and selling to adjust the portfolio to market fluctuations. Mainly, it provided investment banks the opportunity to charge their clients fat fees for the pleasure of putting their expertise to work.

The whole thing blew up as the recent financial crisis unfolded.

It didn't matter how clever or diversified your portfolio was, it lost money.

You could have been in stocks, corporate bonds, government bonds, commodities, in Israel or China or the U.S., or in hedge funds or mutual funds: you lost money. A lot of it, probably somewhere from 30% to 50%.

The failure of diversification can be measured through the correlation between the different investment avenues.

For example, the correlation between stock indices and commodities indices rose to 0.74 last year (1 is 100% correlation). The correlation between oil and stocks rose to 0.7. The correlation between hedge funds and the stock market was 0.5. The correlation between the world stock market and bonds rose to 0.8. The correlation between stocks and real estate funds was also 0.7.

These figures are humiliating to proponents of diversification theory; they illustrate that even diversification could not diminish your risks. There was no escape, no shelter from the crash.

Now the markets are stabilizing. With the panic having abated, the investment management industry is now coming out of its shellshock and searching its soul.

Herd that one before

Part of the debate as it looks back and ponders "what happened" touches on diversification.

Was it wrong all along? Were the financial models that ruled in the last four decades wrong? Most importantly, what next? How should they manage portfolios in this new era?

The easy part is explaining what happened. The explanation is based on two principles: economics and herd behavior.

The economic basis is simple. At the end of the day, all investments are based on the global economy continuing to grow. As long as it grows, all investment avenues profit, generally speaking. Stocks rise, the risk factor in bonds drops, commodities rise, property values increase. This happens simultaneously, everywhere in the world.

The opposite holds true, too. When the global economy contracts, so do all investment avenues. Diversification may help reduce fluctuation of your portfolio, but it can't save you.

The second principle is herd behavior. That's the term used when investors all over the world run to the same investments at the same time. They also dump the investments at the same time. True, there are tens of thousands of investment advisers and funds, but they have their eyes fixed on one another and spend their every effort on being at least as good as the average.

At the end of the day, it's that herd behavior that creates the correlation between the investment avenues.

When a giant hedge fund with tens of billions of dollars under management decides to invest in a "diversified portfolio," it is creating a correlation between the various investment avenues, even if those avenues have nothing to do with one another. Say the fund raises another billion dollars for investment: it immediately raises the asset prices of all the avenues in parallel. When the same fund is hit by a wave of redemptions, then it has the same effect in reverse, depressing asset prices down the line.

It is the herd, and no other, that trampled and destroyed the dividend of diversification. With our own feet, we ruined the mathematical model.

Some in the investment management world argue that the spike in correlation is unusual. It characterizes only times of spiking drops followed by spiking gains, which has been the case these days. The basic logic behind diversification remains intact, they argue. The markets will stabilize and the mathematical models will win again.

Not everyone agrees. If diversification couldn't save an investor from losing 50% of his money within months, then the principle is worthless, they argue. The global economy is just becoming more correlated and the only way to make money is to ride the wave (somehow) - buy at the trough and sell at the peak. According to that logic, only seasoned market animals with their finger on the market's pulse and vast information about companies, trends and what everybody else is doing, can make money.

Is diversification dead? The discussion has only just begun. At this point everybody's groping in the dark and admits as much. Nobody knows how it will end. But there is a loose consensus about a few things.

First: The only thing that can spare investors another crash is super-conservative financial assets, on which, however, they make roughly zero returns. These include cash in the bank and government bonds linked to the consumer price index.

Second: There's no point in expending the energy to constantly twiddle with your portfolio to diversify it. You can settle for gross diversification, for example to place a third of your money in conservative vehicles, a third in stocks and corporate bonds, and a third in mutual funds that do try to time the market.

We'd like to propose a third principle. Make every effort to minimize the fees you pay to portfolio managers, mutual funds, hedge funds and the like. Why? Forking over 2% or 3% of your assets each year, and sometimes more, ruins your returns, in the short run and over time, too. Whether the market is rising or falling, whether asset prices are fluctuating wildly or hovering in place, however much you diversify - those fees will kill you.