Stocks are commonly touted as being safe (real returns greater than 0) over the long run. If you look only at the United States over the past century, this has been the case for any period greater than 20 years. However, in many other countries, stocks needed to be held for a much longer period in order for this statement to hold true. Furthermore, social and economic conditions in the U.S. today are very different from those at the beginning of the last century. In the future, the U.S. may have more difficulty claiming that stocks are safe over the long run - an issue that has important implications for building portfolios in the twenty-first century. In this article, we examine the reasons behind this change and discuss what it means for the equity premium (defined as equity return - cash return). (For further reading, see The Equity Risk Premium - Part 1 and Part 2.)

The U.S. and Uncommon Returns
All too often, investors look at historical U.S. returns as a gauge for future returns. Given that the U.S. saw the largest economic and market growth of any country in the twentieth century, this might not be a prudent strategy, as countries, just like sectors, eventually rotate in and out of favor. Because of its success, however, the U.S. has been the most researched market; more long-term quality financial data exists on it than on any other market. This resulting success bias has strongly influenced investors' projected returns. The bull market of 1982 to 2000 also had an effect on investors' expectations, but given that it was the greatest combined bull market (equities and fixed income) in history, it may be difficult to repeat any time soon. At the beginning of the twenty-first century, valuation levels on stocks were higher and, therefore, not as compelling as they were at the start of the previous century, and investors hurt during the last bubble may be much less likely to pay such high premiums in the future. (For background info, read Digging Deeper Into Bull And Bear Markets.)

Other countries did not share the unprecedented degree of success enjoyed by the U.S. over the twentieth century. According to Elroy Dimson, Paul Marsh and Mike Staunton in their book "Triumph Of The Optimists: 101 Years Of Global Investment Returns" (2002), the U.S. has seen positive real equity returns for every 20-year period in the last century. The same cannot be said for many other markets; countries such as Japan, France and Italy required 50 to 75 years before the same statement could hold true, while the markets of Russia, China and Germany saw total collapse through confiscation, nationalization and hyperinflation in the same period. (See What Is An Emerging Market Economy?) Even though these countries were ravaged by events that were no longer common at the beginning of the twenty-first century, we can't assume that a future global catastrophe (i.e. war, depression or pandemic) could not seriously disrupt the world markets again. (For insight, read The Biggest Market Crashes In History.)

The graph below charts the relative risk and return for various countries in the twentieth century:


It All Comes Down to the Equity Premium
Considering the historical information, what will happen to the equity premium going forward? A strong case can be made for a lower equity premium in the future. Valuations for stocks in the twenty-first century, which far exceed those that existed at the beginning of the last century, provide some evidence of this. For instance, according to Dimson, Marsh and Staunton, the global dividend yield in 1900 was 4.5% - very different from the 1.5% recorded at the end of 2005. Although some of this can be attributed to a shift away from dividend-paying companies, this difference in valuation could explain the high equity premiums realized in the last century, as higher valuations worked their way into those returns. There is also the possibility that valuations could creep back to the lower, historical levels, which would be devastating for future returns.

Statistically, sampling errors exist when return data is limited to only 100 years. For example, there are only five non-overlapping 20-year periods sampled over the last century, and because the data only accounts for one 100-year period, there is no other data to which we can compare it. In other words, 100 years' worth of data may not be enough to be statistically confident about the predictive use of long-term returns from the last century. Finally, proceeding with a lower equity premium and the same historical amount of volatility will mean longer periods before stocks can be considered safe over the long run.

According to Dimson, Marsh and Staunton, the equity premium in the second half of the twentieth century was twice that of the first half. The post-war years saw unprecedented growth and trade in the global economy, while central bankers came to better understand the intricacies of monetary policy. During this time, global investing was a true "free lunch", where you could pick up additional return with little additional risk. This was due to the poor correlation of financial market returns. The financial markets and the world economy were not as integrated as they are today; this has changed dramatically since the 1970s, and global market returns have become more closely correlated. Therefore, this "free lunch" may be harder to realize now. (For further reading see, Broadening The Borders Of Your Portfolio and Investing Beyond Your Borders.)

Some argue that long-term equity returns can even be more certain than bond returns. According to Peter L. Bernstein in his article "What Rate Of Return Can You Reasonably Expect ... Or What Can The Long Run Tell Us About The Short Run?" (1997), stock market returns are historically mean reverting (returns ultimately coming back to their long-run averages), while bonds have not been mean reverting on either a nominal or a real basis. For example, the bond market in the twentieth century was basically two secular bull markets and bear markets - there was no mean reversion about it.

Despite this argument, the yields on inflation-protected securities (another basis for calculating equity premium) were actually dramatically compressed at the beginning of this century, as a glut of global savings resonated throughout the world. Since that time, the world has generally become a much safer place to invest. Shouldn't this relative safety be reflected in lower equity premiums required by investors? At any rate, it certainly helps to explain the increase in valuations in the last half of the twentieth century.

Assuming a lower equity premium in the future, people will need to change the way they invest. Personal savings will have to increase to offset lower real returns and real assets may offer a more viable alternative to financial assets. Finally, sector and stock selection will become even more important.

In terms of market returns, the twentieth century was more favorable to the U.S. than to most other countries. For these countries, the risk-return tradeoff for stocks has not been as attractive, and it is more difficult for them to make well-supported claims about the success of holding stocks long term. The conditions that existed for the U.S. at the beginning of the last century did not exist at the beginning of this century. This may mean a lower equity premium, requiring changes to how much you save and how you invest your money.

To read more, see The Stock Market: A Look Back.

Want to learn how to invest?

Get a free 10 week email series that will teach you how to start investing.

Delivered twice a week, straight to your inbox.