From the January 2009 issue of Research Magazine • Subscribe!

January 1, 2009

Theorizing Risk with Harry Markowitz

Harry Markowitz is best known for his pioneering work in Modern Portfolio Theory (MPT). He evaluated the impact of things like asset risk, diversification, and correlation on the expected return of an investment portfolio.

His cutting-edge research changed our perception of risk and return along with revolutionizing the way investors build portfolios.

In 1990, Markowitz was awarded a Nobel Memorial Prize in Economic Sciences and he serves on the advisor panel of investment research firm Research Affiliates. In addition, 81-year old Markowitz still works as an economist at the Rady School of Management at the University of California, San Diego. He took time to visit with Research magazine to offer his insight about the current state of the financial markets and investing.

With the stock market down by more than 50 percent, a deflated housing market, a weak job and economic climate, when was the last time you can remember this many things going wrong all at the same time?

Never. Remember I was born in 1927, so I wasn't aware of 1929 through 1933. I think the big difference between now and say the tech bubble bursting is that it affected all of those folks that were foolish enough to concentrate their portfolios in high-tech stocks. That subsequently led to a recession which I think will prove to be mild compared to the current recession. The closet thing I can think of that's similar is the Japanese bubble that broke shortly after 1990 and was led down by real estate. If we play our cards wrong, we're in for a decade of stagnation.

What do you think is the basic cause of our current economic problem?

For a good many years Congress has been pressuring Fannie Mae to increase the mortgages they hold or buy for low income housing. I remember an interview with the former head of Fannie Mae and he said, as early as 2000, by 2010 there would be $2 trillion worth of low income housing. And then around 2005, there was considerable pressure by Congress and regulators to increase the amount of low-cost housing. As it turns out, the only way this could be done was by lowering borrowing standards and once you've done that all sorts of people can get loans. Somebody's who's reasonably wealthy can pay no money down for a second home and if the market goes up they won; if the market goes down they just put it to the bank.

How would you grade the U.S. government's response to the financial crisis? What should they be doing?

I think it's a sad situation. First, Congress should announce to Fannie Mae, to themselves and to the world that whenever there's a conflict between the objective of low-income housing and financial stability, the latter has absolute priority. Second, the government should do a survey of what goes into what. For example, they need to determine what mortgages go into a CMO or what tranches of a CMO or anything else go into a CDO and so on. This is a big job, but no bigger than something like the annual survey of manufacturers which the Census Bureau does every year. This is a lot more urgent than that. Then the government needs to use math and computers to figure out what the direct and indirect exposures are of this paper. This math is also similar to the math used by search engines. After that's done, the data needs to be aggregated by Zip code and payment history and then distributed to relevant parties like stockholders, counterparties, regulators and supervisors. Ideally, the survey should be updated daily like a clearinghouse.

A lot of blame for the current financial mess has been put on the lack of transparency with complicated financial products. Wall Street's propensity to innovate has gotten not just itself in trouble, but the global economy. Any comments about that?

Nobody knows the direct and indirect exposure of various kinds of financial paper. Mortgages are pooled together into a collateralized mortgage obligation that's split into tranches and various people can invest in it. For example, a collateralized debt obligation could have as one of its assets a tranche from a CMO, or tranches from CDOs, which can go into other CDOs and so on, until nobody knows what anybody has. If A depends on B and B depends on C and C depends on D, then if A gets a cold, everybody sneezes. We've always had contagion because if somebody has problems financially, everybody gets more cautious and they try to spend less and, as they do that, the person downstream has less income as a result.

What about other complicated financial instruments like credit default swaps?

I think credit default swaps are called that rather than insurance to confuse insurance regulators into thinking they don't need reserves, but they need more reserves than life insurance. Deaths are fairly uncorrelated events whereas business risks [are] fairly correlated risks. If General Motors is doing bad like it is doing now, then Ford is probably going to do bad like it's doing now and so are their suppliers. If you have uncorrelated risks, then by diversifying sufficiently you can drive a risk that's as close to zero as you want, but that is not true with correlated risks. And that, by the way, is the same information from Chapter 5 of my 1959 book.

I don't want the financial system to collapse, but I have absolutely no sympathy for the management that said to their financial engineers, "If you say these things are OK, then they must be OK." I hope stockholders diversified.

What are some things investors can do to reduce their financial risk -- without reducing their returns? Is that possible or is it just a pipe dream?

Yes, it's a pipe dream. MPT distinguishes efficient and inefficient portfolios. Efficient portfolios have minimum risk for a given expected return or maximum expected return for a given risk. If you can increase expected return without increasing risk then you must have an inefficient portfolio. Once you have a portfolio that gives you an efficient combination of risk and return, you cannot get more return without taking on more risk and you cannot lower risk without giving up return. If you could, then the portfolio wouldn't be efficient.

What about leverage?

Traditional applications of MPT by pension funds and endowments usually use a 60/40 or 70/30 mix of stocks, bonds and a little cash. Places like AIG figured out a way to leverage much more than 2-to-1 and they are the ones that have gotten into trouble. Suppose there's a pension fund with $6 billion [in] stock and $4 billion in bonds and if the $6 billion has gone down to $4 billion, it hurts, but you're still alive and in the long-run you'll do just fine. But if you're leveraged 10-to-1 and you're mark-to-market, you're dead.

In the indexing world, there's been a rather heated debate about the validity of market-cap-weighted indexes versus fundamentally weighted indexes (you call them "efficiency weighted portfolios"). Can we really say fundamentally weighted portfolios are indeed "indexes," or are they just actively managed funds in disguise?

If everybody knew the true value of a stock, whatever that means, then all efficient portfolios would be the market. But if you assume that there is a true value but that the market is sometimes above it and below it, then it turns out that a fundamental index will outperform a cap-weighted portfolio.

As you know, many fundamentally weighted portfolios are constructed with a bias towards small-cap stocks and value companies. Even though these two particular biases have produced historical returns that have beaten the market, what if they don't repeat their success?

In the old days, there were just two asset classes: stocks and bonds. Then they started classifying stocks by large-cap, mid-cap and small-cap as well as by growth and value. Sometimes large-cap does better than small-cap and sometimes value does better than growth. It's true that on a forward-looking basis we don't know if it'll be a large-cap year or small-cap year or a growth year or value year. But the reason fundamental indexes tend to be smaller-cap than a cap-weighted index is because I would argue that cap weighting automatically overweights anything whose cap value is too high as compared to its true value.

What are some fundamental mistakes that investors make -- ones that you've observed -- that they consistently repeat?

Active investors like day traders and passive investors gross the same amount, but the brokers of the active investors get rich. The smart investors put their money into their retirement account and they don't chicken out just because the market drops in one year. The dumb ones buy when the market is up and they think it's going higher and they sell when the market is low and they think the market is going lower. And then of course, there are professionals and amateurs that leverage too much.

Exchange-traded funds are low-cost index funds that trade like stocks and, compared to closed-end funds and traditional mutual funds, they are the "new kids on the block." What do you think about ETFs?

They're great. I have them. Something like the SPDRs has tremendous trading volume and is very low in cost. Other ETFs are higher-cost. I believe individuals should have diversified equity portfolios but not be 100 percent in equities. But how does a person accomplish that goal without the high front-end load or high ongoing expenses? They do it by owning an inexpensive mutual fund or an ETF.

Computers and the Internet have made it easy to access lots of information. Compared to previous generations, do you think the investing public is smarter?

What we have today is lots and lots of data that we didn't have 20 years ago and the ability to easily access it. It's like supplying someone with a very sharp knife. They can do surgery or they can cut themselves. The ability to trade fast anytime you want is a great facility if you use it just once in a while. To brag a little bit, when Bear Stearns first hinted at its problems, about a week later I ordered my broker to sell some of my ETFs and they said it would be executed the next morning. It was convenient. On the other hand, it invites people to trade and cause lots of additional costs for themselves. People still have to distinguish between data and truth. They need to get a combination of stocks and bonds they can live with.


Ron DeLegge is the San Diego-based editor of

Reprints Discuss this story
This is where the comments go.