Less than a quarter of Americans get the most important investment question right

Don’t feel bad if you didn’t get the answer right. You’re in good company.

Recently, Gallup asked Americans what they thought their best investment bet was over the long run. 24 percent of Americans named stocks and mutual funds — but the same share named gold, and even more (30 percent) named real estate. The trend lines are actually positive, as in 2011 a baffling 34 percent of Americans named gold as their top pick:

There’s a right answer here — and it’s one that about two thirds of respondents who answered the question got wrong. If history is any guide, stocks are the best bet in the long run, and gold and real estate certainly are not.

Wharton professor Jeremy Siegel’s popular book Stocks for the Long Run (now in its fifth edition) illustrates this point very effectively using historical data Siegel compiled encompassing the whole 19th and 20th centuries through the present. While for the first part of the 19th century, American stocks and bonds were roughly equivalent in performance, stocks have pulled away ever since.

Siegel helpfully summarizes his findings in the chart below. Keep in mind while reading it that, unlike the other charts in this post, it’s logarithmic rather than linear. Stocks in 2012 are a little less than twice as high up the chart as bonds, but they did way, way, way better than twice as well. If you invested $1 in stocks in 1802, you’d have $704,997 (after adjusting for inflation) in 2012. If you invested $1 in bonds, you’d have $1,778. That’s a two order of magnitude difference, not a factor of two difference:

Gold has never even come close. Similarly, real estate has generally generated returns in line with inflation — far below the returns to stocks or bonds. Nobel-winning economist Robert Shiller once noted that from 1890 to 1990 (before the ‘90s/00s real estate bubble), the appreciation in housing in the US was roughly zero. “If you think investing in housing is such a great idea, why not invest in cars?” he elaborated. “Buy a car, mothball it, and sell it in 20 years. Obviously not a good idea because people won’t want our cars. It’s the same with our houses. So, they’re not really an investment vehicle.”

Stocks’ advantages hold up even if you focus in on recent decades. While the stock market took a beating in 2008, while bonds proved more resilient, the investment firm BlackRock found that you’d still be better off investing in stocks from 1994 to 2013. It didn’t make a huge amount of difference whether you chose “small cap” stocks (shares in firms worth between the hundreds of millions to low billions, and represented here by the Russell 2000 index) or “large caps.” That said, “growth” large caps, which cost more relative to company earnings but have higher forecasted growth potential, underperformed compared to “value” large caps, which have worse growth forecasts but better earnings ratios.

A diversified portfolio, with 35 percent of assets in bonds and 65 percent in various kinds of stocks, performed worse than any pure-stock portfolio, but much better than bond (or “fixed income”) investments or international stocks. All told, a diversified portfolio did 60.5 percent better than a bond portfolio, and an investment in the S&P 500 (“large cap core”) did 91 percent better:

It’s important, however, not to interpret this as a sign that your portfolio now, forever, and always should be mostly or solely in stocks (and it certainly doesn’t mean you should be buying and selling individual stocks, but that’s a matter for another post). A heavy weighting toward stocks makes sense for young investors who want to maximize savings 50 to 60 years in the future; in that case, long-run returns are the most relevant factor to consider. But the closer one gets to cashing out investments, the more volatility starts to matter. Knowing that the S&P 500 outperforms bonds in the long-run is small comfort when the economy collapses right before you retire and your S&P investments lose 56.8 percent of their value in little over a year.

Siegel has another chart illustrating this point well. It shows the best and worst performances for stocks, bonds, and short-term Treasuries over different periods of time. For short periods (less than 10 years), the worst performance for stocks is below the worst performance for bonds, and the best performance for stocks is higher than the best for bonds; over the short-run, bonds are a safer bet. But for periods of 10 years and more, the worst performance by stocks is actually better than the worst by bonds, even as the best performance remain higher than the best bond performance. Stocks become both a better investment and less risky than bonds over a long enough time span:

Does this mean that there will never be a long-run period over which bonds outperform stocks? Nope; indeed, Siegel notes that bonds narrowly edged out stocks from the beginning of 1982 to the end of 2011. But that’s a fluke attributable to both the extremely high bond interest rates of the early 1980s and the crash of 2008. Before that, the last time bonds beat stocks over a 30 year period was from the start of 1832 to the end of 1861.

And those trends do suggest that, in the short-run, it’s much more reasonable to move toward bonds. As a table of Siegel’s shows, over time the probability that stocks will underperform bonds shrinks dramatically. It doesn’t hit zero — stocks won 99.3 percent of 30 year intervals from 1871 to 2012 — but it gets very close. But that also means that there’s a reasonably high chance bonds will out-perform over a one to five year interval:

So it makes sense to move into lower-risk, lower-return investments the closer you get to wanting to cash out.