Chances are that if you manage investment portfolios, modern portfolio theory (MPT) lays the groundwork for how you do it. It provides a framework for advisors to help their clients find the best return on their money with the least amount of risk. Conceptually, it allows them to do so by examining the effects of asset risk, correlation and diversification on their portfolio returns and then by prompting them to select a portfolio accordingly.

But this landmark theory of investing was put to the test in 2008, which saw the S&P 500 index drop 38%. So does modern portfolio theory hold up?

The answer is yes, though much depends on how you split up your clients' investment pie, says Harry Markowitz, who won the 1990 Nobel Prize in economics for developing Modern Portfolio Theory in 1952 and 1959. "It is sometimes said that portfolio theory fails during a financial crisis because all asset classes go down and all correlations go up," the 82-year-old Markowitz acknowledges in a telephone interview from his office in San Diego.

But this, he stresses, was already predicted by the theory. "Generally, asset classes move roughly in proportion to their historical betas."

Markowitz teaches the landmark theory at the Rady School of Management at the University of California, San Diego. He's also working on a new book and consults with companies such as investment advisory firm Index Fund Advisors and Guided Choice, an advisor to 401(k) plan sponsors.

When he goes for his daily lunchtime walks, Markowitz notices license plate numbers, and routinely calculates the probability of seeing the same number reoccur. For example, the probability of seeing a license plate with four of a kind such as 6666 is one in one thousand, he says. But every year, if you walk enough, you will see another license plate with four of a kind.

It's the same, he says, with the stock market.

Markowitz advocates optimizing portfolios to get the best return per unit of risk by using asset correlations dating back to 1926, rather than using short-term and current correlations. History tends to repeat itself, he believes, but not in the same sequence. So there is always a chance the investor could experience losses similar to those in 1929, 1987, 2002 and 2008.
"Personally, I think that nature draws from the basket known as the S&P 500 randomly every year," he says. "The stock market losses in 2008 have a probability of occurring about once out of every 40 years. The 38.5% loss on the S&P 500 was more than 2.5 standard deviations below the mean."
Yes, the 2008 market crash could happen again, he says. But you don't know when. The losing cards are in the deck.

Markowitz's theory for allocating assets has been refined by subsequent research. The capital asset pricing model, for instance, designed by William Sharpe in 1963, gave money managers an extra measure of risk to consider. Sharpe's theory considers systematic risk, or how an asset moves in relation to the overall market. Under Sharpe's theory, systematic risk is measured by a portfolio's "beta" value. The market, as measured by the S&P 500, has a beta value of 1. An asset with a 1.2 beta value is about 20% more volatile than the market. By contrast, an asset with a 0.8 beta is 20% less volatile.

Additional refinements were devised by Eugene Fama and Kenneth French in 1995. They fine-tuned the model to account for investment styles, taking into account whether the stock was a large, medium-size or small company, and whether it was a growth or value play. So to reduce volatility today, advisors often use modern portfolio theory to construct portfolios and incorporate a wide range of United States and overseas asset classes.

Markowitz says that financial advisors concerned about re-experiencing recent large market losses should optimize their clients' portfolios by removing some risk from the investments and giving up some of the return. Such portfolios would have lost less in 2008, and they would have hit their break-even point in a shorter period.

For example, the S&P 500 lost 38% in 2008 and higher-beta emerging markets indexes fell 54%, he says. Corporate bonds might have fallen in value, but they declined much less than stocks, and government bonds even rose in value (the five-year government bond index rose 8.4%). Small capitalization stocks dropped in value, but not as much as expected, based on their standard deviations. Meanwhile, large capitalization stocks performed worse than expected. Taking all these factors into account, a simple fifty-fifty split between the S&P 500 and the Lehman Brothers Government Bond Index would have lost just 12.5%.

Markowitz, by the way, says he has about 60% of his own portfolio invested in exchange-traded stock funds more heavily weighted toward small companies and emerging markets. The rest is invested in short-term and intermediate-term bonds-not bond funds, which he believes invest in "too much crap." This may sound a little risky for someone his age, but he says, "I feel comfortable and have enough money in municipal bonds so that if I die my wife has enough to live off tax-free income. I can afford to be more in equities."

Over the long run, modern portfolio theory helps advisors spread their clients' risk and build wealth. Nevertheless, the theory is not without critics. A chief complaint is that Markowitz's model assumes asset returns are normally distributed. Frequently, however, stock market returns are not normally distributed. There can be large swings of three to six standard deviations from the mean, and these occur more frequently than they should if the returns were normally distributed. Other research shows that the capital asset pricing model can't always be explained by a portfolio's beta value. Low-beta stocks sometimes deliver higher returns than high beta stocks.

Markowitz disagrees with the critics, saying his model never assumed that the probability distribution would be normal. His research shows, rather, that mean variance portfolio relationships are a good approximation of the expected value of a portfolio.

"It is another myth," he says, "that you can't invest in assets [using modern portfolio theory] that are not normally distributed."
For example, he stresses that derivatives, such as stock options, might be optimized in a portfolio to improve risk-adjusted rates of return as long as a financial advisor has good data on the covariance relationships to other portfolio assets.

Markowitz advises financial advisors to estimate the likely returns, risk and correlations among various asset classes and use these inputs to conduct a modern portfolio theory analysis. "This produces a curve of 'efficient' risk-return combinations. If you want greater return on average, you have to take on greater risk. If you want less month-to-month and year-to-year fluctuations, you have to accept less return on the average in the long run."

It's better to diversify across asset classes as well as within the portfolios, he says, which is more "efficient" than doing only one of these, or neither. For example, riskier portfolios may contain greater weightings in emerging market and small company stocks, based on investment style. Less risky portfolios might be weighted more toward large-capitalization U.S. and foreign stocks, based on investment style, as well as short-term and intermediate-term bonds.

He emphasizes that the shorter the investment horizon, the greater the risk of losses because there is fatter tail risk. But the longer you hold asset classes, the greater the probability the distribution is normal. As a result, two-thirds of the time, returns typically fall between negative 10% and positive 30%.

Be cautious about investing in alternative assets, ETFs that invest in alternatives, private placements and commodities, he warns, unless you have the time and expertise to value them. Those types of investments, he says, are best left to people like Warren Buffett and David Swenson of Yale University's endowment fund.

Also, he says, don't put too much emphasis on Monte Carlo simulations, which can scare clients by showing them, for instance, that one out of 20 times, the money doesn't last as long as they do. On the plus side, the simulations often inspire 401(k) defined contribution plan participants to save more for retirement.

Markowitz cautions against "model risk." If you are using a specific type of investment model that tells you how to invest using fundamental, economic and/or technical data, the model may ignore or downplay the possibility of large losses that have a 1-in-20 or 1-in-40 chance of occurring, he says.

It's also important, he says, for financial advisors to identify their clients' risk exposure using five dimensions: their time horizon, their liquidity needs, their net income, their net worth and their investing knowledge and attitude toward risk. Educate clients about the risk exposure of their portfolios, he suggests.

"At any point in time, we look back at the past, and make our estimates and decisions for the future," Markowitz says. "The future is always uncertain. We should make our best estimates for 'the next spin of the wheel,' and then choose an appropriate point from the implied risk-return trade-off curve.

"Depending on our risk capacity, perhaps we will select a more cautious portfolio, loaded with lower beta securities or asset classes; or conversely, we will choose a point higher on the frontier, with higher yield, but with higher beta securities or asset classes.

"If the market goes up dramatically, those with high beta portfolios will be happy; if it goes down, they will be sad. You pays your money and you takes your choice!"