Stocks For The Long Run? There Is No Long Run

Last Updated Oct 12, 2009 1:07 PM EDT

The author and promoter of Stocks For The Long Run, Professor Jeremy Siegel of The Wharton School, is back. A few days ago in an op-ed submission Siegel revved up his old hypothesis - that investing in stocks always beats investing in bonds, sort of. In my view, his advice for individual investors was simplistic and dangerous when it was fresh in 1994, and seeing that Dr. Siegel's patter has not been informed by the two stock market crashes since then, the message has become only more so. (This is a long post, but worth it; please bear with me.)

I. The beating stocks took in 2008 and 2009 did plenty to disprove, or at least soften up, Siegel's hypothesis. At the stock market low in March, "stocks for the long run" (hereinafter SFTLR) was in tatters, because at that point, the returns to U.S Treasury bonds had beaten equities for the prior 40 years. (If 40 years doesn't constitute the long term, I don't know what does.)

Therefore this week I was disappointed to see that the Financial Times, a publication that I adore and sometimes have the privilege of writing for, had given Dr. Siegel time on its podium. Here's a sample of his defense of SFTLR:
[F]or the 13 10-year periods of negative returns stocks have suffered since 1871, the next 10 years gave investors real returns that averaged more than 10 per cent per year. This return has far exceeded the average 6.66 per cent real return in all 10-year periods, and is twice the return offered by long-term government bonds.

Strong future returns also followed poor returns if one extends the analysis to the worst-performing of all 127 10-year stretches since 1871. Without exception, for each 10-year return that fell in the bottom quartile, the following 10-year period yielded positive real returns and the median return exceeded the long-run average...
He went on to suggest that the comparison with bonds for the last 40 years wasn't fair, because their returns had been above average. Huh? He didn't omit the above-average years for stocks.

And he counters:
In fact, with the recent stock market recovery and bond market decline, stock returns now handily outpace bond returns over the past 30 and 40 years.
To my thinking, the question is not how you parse "127 10-year intervals" this, and "bottom quartile, real return" that. Instead: Do stocks always (or nearly always) beat bonds? And is that over the investment horizon of real people? If the answer is no, and it's instead a matter of cycles, then whether stocks or bonds outperform depends on how it's measured.

I'm not the first to criticize Seigel's methods and conclusions. One recent refutation, from excellent investment writer Jason Zweig, called the professor out for basing his long run, which goes all the way back to 1802, on shaky research conducted by a few professors in the 1930s.

II. Many years ago I went to business school. Of all the classes I sat through, I remember only a few constant principles. Here's one, from a class on portfolio theory: "The long run is a series of short runs."

Since then I have devised a corollary: "Individuals don't invest over the long run. Each has one run, over a fixed time frame."

You save a certain amount, and have to invest it in the markets that come your way. If you're lucky, you'll have good markets in your mid-to-late working years, when you've built up a lot of capital and can double your money. If not, maybe you can hang on and work a few more 10-year intervals, until Dr. Siegel's long run prevails.

III. Siegel's closing comment in this week's apologia for SFTLR says it all:
The recent behavior of stock market prices sheds some light on a phenomenon which has long puzzled economists: why do stocks over the long run yield so much more than bonds? The pain that investors often suffer, such as in the recent bear market, forces many to forsake equities altogether. This drives stock prices down and enhances their future returns. Equities offer investors excellent returns to those willing to accept the market's volatility.
So the reason stocks outperform bonds is not that they do so every year, or in most years, but that they go down a lot? It's a cycle, a game of musical chairs. But that means investing in stocks is a good idea sometimes, and a lousy idea at others.

OK, you'll have to wait a while. But the idea that stocks always win if you stay in the game long enough is not relevant or realistic either. Real people don't have infinite wealth and open-ended working lives. And the way market cycles interact with their working years and patterns of building assets adds an important additional dimension.

You don't have "the long run." You have one run, over your working years. Be sure you understand what that means, and don't let a book title, even a professor's, determine your investment policy.

Financial disclosure: The writer owns a lifetime supply of stocks, but plenty of bonds too.
  • John Keefe

Comments

Market Data

Watch CBSN Live

Watch CBS News anytime, anywhere with the new 24/7 digital news network. Stream CBSN live or on demand for FREE on your TV, computer, tablet, or smartphone.

Market News

Stock Watchlist