Courtesy of Knowledge@Wharton

Stock market investors have suffered deep losses in the past 18 months, challenging the belief that stocks are the best long-term investments. Indeed, Wharton finance and economics professor Robert Stambaugh recently coauthored a paper titled Are Stocks Really Less Volatile in the Long Run? which suggests that equities are subject to bigger price swings than previously understood. The research adds a new perspective to the work of Wharton finance professor Jeremy J. Siegel, author of the book, Stocks for the Long Run, which says stock returns more than offset risks if you stay with the market through its ups and downs. In a recent interview with Knowledge@Wharton, the professors described their views about the market's long-term behavior.

An edited transcript follows.

Knowledge@Wharton: Professor Stambaugh, your work finds that volatility in stocks -- the ranges of ups and downs -- can be considerably more severe than most people believe. And your paper focuses on uncertainties that cloud the future. Can you put this in laymen terms for us?

Robert Stambaugh: Generally when we think about volatility in the stock market, we think about the value of the market fluctuating, typically around some sort of trend or long-term expected rate of return. Our work makes the point that uncertainty about that trend itself adds to the uncertainty that investors face and should be perceived by them much like volatility -- much like the fluctuations around the trend. That uncertainty about the trend itself becomes more important the further into the future you project investment outcomes. So our paper basically makes the point that to an investor with a long horizon, stocks actually are riskier per period. That is, the rate at which risk grows over the horizon such that it makes the investment riskier over the long run. This is basically in contrast to what we think of as more conventional wisdom that says over the long run, fluctuations in the stock market will to some degree cancel each other out and, therefore, make risk to an investor in the long run look [smaller] per period than [it would] to a short-run investor.

Knowledge@Wharton: There are a number of factors that you say contribute to this uncertainty about the future. What are the most important of those?

Stambaugh: Probably the most important is just uncertainty about the overall long-run trend. The other feature of the stock market that contributes to uncertainty is the fact that at some points in time we think the expected rate of return is higher than at other points in time. In other words, over time the rate of return that you can expect to earn over the short- and intermediate-terms fluctuates. The fact that the expected return fluctuates also adds to uncertainty because we do not know -- for example -- if expected returns are currently high, which many of us would guess they are. We don't really know how long they're going to stay high. That may in part depend on how quickly the economy recovers. That additional uncertainty also contributes to higher uncertainty in the long run.

Knowledge@Wharton: What would be an example of a kind of event that could occur that people simply can't take into account ahead of time? I saw in one story a mention of global warming, for example.

Stambaugh: Global warming is interesting. It provides an interesting analogy to this concept because ... we might be very uncertain about how quickly the Earth is warming, but that uncertainty doesn't much impact our uncertainty about crop output and economic output next year. But if we look 50 years down the road ... uncertainty about the rate [at which the Earth is warming] has a much bigger impact on our overall uncertainty.

Knowledge@Wharton: Professor Siegel, your work in Stocks for the Long Run, and in the years since then, talks a lot about the long-term trends -- a couple of hundred years of stock market data, and bond and cash returns -- [showing] that volatility does tend to even out over time. Is that right?

Jeremy Siegel: Yes. My empirical work -- which tracks stock returns since the beginning of the 19th century, so now we have a little bit more than 200 years of data -- showed that stock returns display what economists call "mean reversion," or reversion to the mean. In the short run, there seems to be a lot of volatility, uncertainty. But if you draw a trend line... it tends to fluctuate around that. And if I understand Rob's work, he still agrees there is mean reversion.

Stambaugh: Yes, exactly.

Siegel: He's not denying it and most researchers -- actually even before me -- had suggested [mean reversion]. When I did it, it got stronger. And almost everyone has confirmed it. So the mean reversion story, I think; it is still there. What Rob is bringing up is, as he calls it, trend uncertainty. That analogy to global warming is a good one in the sense that ... crop yields fluctuate a lot year-to-year. But we know that over 5- or 10-year periods they tend to get back to a mean. But if there's going to be some big change in the future -- and what Rob seemed to say is even with 200 years of data -- you can't really be that certain of the long-term trend. In my first chapter in Stocks for the Long Run ... I say that we could be seeing that the last 200 years are the golden age of capitalism. When we look through the centuries, there have been an awful lot of huge changes. That golden age may be over. So, even 200 years doesn't give you confidence. We have empires that have ruled 1,000 years and more and then they fail. That sort of big picture uncertainty really could be a dominant feature looking far into the future.