Gold and Stocks: A Prediction -- and a Challenge

Last week, Gallup released results of a nationwide survey asking Americans what they think is the best long-term investment. The results weren't surprising:

• 34% said gold.
• 19% said real estate.
• 17% said stocks.
• 14% said savings accounts.
• 10% said bonds.

Similar surveys have been run over the past decade. Stocks are usually near the top. Gold usually isn't even an option. Digging through news archives, I couldn't find a single survey asking the public what they thought about gold between 1998 and 2002 -- anywhere. People just didn't seem to care. Today, Internet tracking services like this tell you most of what you need to know.

I'm going to make one prediction and pose one challenge to readers. My prediction is that over the next 10-15 years, reality will prove today's expectations of the best long-term asset to own painfully wrong. My challenge is for readers to find a historic example of that not ever being the case.

There are only a few ironclad rules of investing. One is that there's a negative correlation between sentiment and future returns. When the public expects outsized returns, they practically guarantee otherwise. When people won't touch an asset because they think it's toxic, the stage is set for outperformance. Think about gold. The idea of investing in gold was virtually unmentionable 10 years ago. Among other things, that set the stage for an epic rally ever since.

Other recent examples of assets doing the opposite of expectations are powerful.

• In 1999, a survey by the Securities Industry Association showed that investors expected stocks to return 30% a year going forward. Ever since, they've returned roughly 30% less than that.
• As Foolish colleague Todd Wenning wrote last week, BusinessWeek penned a cover story headlined "The Death of Equities" in 1979. Reality was the opposite: 1979 was close to the birth year of a two-decade mega-rally.
• In 2005, just 5% of homeowners thought the value of their home could decline. Another 2005 survey showed that the region most bullish on housing was Florida -- consequently, it was among the hardest hit once the bubble burst.

Run through history, and you see the same pattern over and over again. Assets investors are the most bullish on perform miserably in the future, and vice versa.

The reason why isn't a mystery, though this is a case where psychologists can explain what economists cannot. One of the most powerful forces in behavioral finance is called recency bias. It states that when we try to predict the future, we tend to just extrapolate the recent past. If an asset did well in recent years, our expectations of future returns jump. If it crashed in recent years, we give up on its future.

That tendency to chase patterns is natural, but history nearly always plays out the other way. Since 1820, the real (inflation-adjusted) returns of major assets been good for stocks, OK for bonds, and near zero for gold and real estate.

Those returns have strayed dramatically over the past 10 years -- particularly for stocks and gold:

Source: Deutsche Bank Long-Term Asset Return Survey.
*Some studies find the long-term real return of real estate is zero. Part of the positive long-term return shown here is likely due to changes in average home sizes.

Here's what's important. If one breaks up these 190-year returns by decade, an unmistakable pattern emerges among all asset classes: Periods of above-average returns are invariably followed by periods of below-average returns, and vice versa. This isn't surprising, since there's a negative correlation between sentiment and future returns, and sentiment is skewed by recency bias.

The fact that recently stocks and gold have performed abysmally and incredibly over the past decade does a number on our expectations of future returns. But historically, it makes the odds of repeating that performance over the coming decades extremely low. Since 1820, gold has produced two consecutive decades of positive real returns only once. Stocks? They've never undergone two consecutive decades of negative real returns. When there is a decade in the red, the preceding 10 years have historically brought stock returns of double the average.

If history is any guide, stock returns over the next 10-15 years will be above average, and gold returns will be below average -- just the opposite of what the Gallup survey predicts.

So here's my challenge. Can anyone think of a time when an asset the general public was most bullish (or bearish) on performed the best (or worst) over the subsequent 10 or 15 years? I can't. If one exists, share it in the comment section below. If not, think long and hard about how you feel about stocks and gold.

Check back every Tuesday and Friday for Morgan Housel's columns on finance and economics.