Historical Performance and Future Stock Market Return Uncertainties

by Ed Easterling

There has never been another period in history more demanding of investor efforts, nor as rewarding of investor diligence, as today’s. The current environment poses a broad range of uncertainties. Investors are now confronted by strong contrasts between conventional wisdom and unconventional insight. For example, conventional wisdom points to historical average returns for long-term investors, but unconventional insights from history tell a different story.

“Long-Term” Returns

The well-known and most-respected long-term average statistic from the stock market is the one provided in the annual compendium published by Morningstar “Ibbotson Stocks, Bonds, Bills, and Inflation Classic Yearbook.” Consultants, financial analysts, advisers and investors seeking a benchmark for long-term stock market returns reach for this annually revised testament of historical financial market data. [Editor’s note: This book can be found at many public libraries.]

The number that everyone seeks in the book is the long-term average return from the stock market, which is near 10%. Conventional wisdom believes that any number representing more than 80 years of history must be a valid indication for the next decade or two. Everyone knows that the stock market has ups and downs, yet 80 years is certainly enough time to produce a valid measure for the average condition, right? Well, actually not.

The three components of stock market returns are dividend yield, earnings growth, and the change in the price-earnings ratio; any other elements will fall within one of these three components. The discussion of the historical data in the Ibbotson yearbook provides the details for all three components and the insights needed to fully appreciate the “Average” (capitalized to acknowledge its aura as a presumed market truism).

The Ibbotson yearbook is updated annually to include another year’s data in the long-term series. To reduce the distortions from any single set of data, Figure 1 reflects the average of multiple series for greater consistency. From each of the three yearbooks preceding 2010, the long-term total return from the stock market is near 10%. The components vary slightly across the three publications, yet the total return is fairly consistent. Earnings growth contributed between 4.1% and 5.3% toward the total return, dividend yield provided between 4.3% and 4.4%, and changes in the price-earnings ratio added 0.6% to 1.3%. Simplifying the ranges, the average values are 4.7%, 4.4%, and 0.9%, for a total of 10%.

Source: “2011 Ibbotson Stocks, Bonds, Bills and Inflation Classic Yearbook” (Morningstar, 2011).

According to the trite expression, however, the devil is in the details.

Most noteworthy, the series begins in 1926. This is critically important because the series starts during a period when the market’s price-earnings ratio was fairly low. Ibbotson’s price-earnings ratio for 1926 is 10.2. Since the level of the price-earnings ratio is such a major driver of stock market returns, the starting level of the price-earnings ratio for this most recognized series is an under-appreciated footnote. If the series had started with a price-earnings ratio closer to the average for the recent decade of nearly 25 (or even at recent levels near 20), then the accepted historical average return would be closer to 7%!

When assessing the contributions of the components to the long-term average, it is worth noting that a starting price-earnings ratio near 20 results in a dividend yield closer to 2.5%. [Editor’s note: Valuations and yields are inversely related when the dividend is unchanged. A stock’s yield is calculated by dividing the four-quarter dividend by the share price.] The level of the price-earnings ratio is a major driver of the dividend yield investors receive. When the price-earnings ratio is near 10, the dividend yield is twice as high as it is when the price-earnings ratio is 20. That element alone strikes almost 2% from the long-term average return.

Further, the long-term average includes a near doubling of the price-earnings ratio. From current levels, virtually everyone accepts that a doubling in valuation is unlikely, and many are concerned that the market’s price-earnings ratio may decline from current levels. For this assessment, assume that the price-earnings ratio remains unchanged at currently high levels. [Editor’s note: This article was written prior to this summer’s correction.] By excluding the effect of an increase in the price-earnings ratio, the long-term average return is reduced by another 1%.

Taken together, the two issues of dividend yield and price-earnings ratio change explain the difference between a long-term average return of 10% and one of 7%. Had Ibbotson started the long-term series years earlier or later, when the price-earnings ratio was closer to 20, all of us would carry quite different expectations for long-term returns.

Therefore the long-term return of almost 10% represents the return for an investor who enters the market when the price-earnings ratio is 10.2, stays for more than 80 years when the inflation rate averages just over 3% (to boost nominal earnings growth), and exits after inflation subsides and the market price-earnings ratio has just about doubled from where it started.

Just like the favorite shirt or outfit you own that you hope to keep long enough for it to come back into style, the long-term average return of 10% will be a relic in the back of the closet for any period that does not start from a low price-earnings ratio. When starting from a relatively high price-earnings ratio, two of the stock market return components cannot contribute as much as they did from 1926.

Decades, Not Centuries

Some people will identify with the long-term view; others will be curious to see the effect that these factors have on their own investment horizons. Ten years is generally considered long enough to smooth the yearly ups and downs in the market. Many people expect that the cumulative return over 10-year periods will begin to concentrate near the long-term average return of 10%.

There have been 102 rolling 10-year periods since 1900, covering more than a century of stock market history. Of course, very few, if any, 10-year periods will have total returns (including dividends) of exactly 10%. Yet, many investors and advisers either implicitly or explicitly assume that stock market returns for this decade will be inside the 8%–12% range. This is especially true considering that the assumption by many investors for overall investment portfolio returns is 8% or more.

Consider that the expectation for returns from a portfolio includes a blending of assumptions, combining relatively higher stock market returns with much lower bond investment returns and, for some investors, returns from alternative investments. For the blended portfolio to generate overall returns of 8% after expenses, stocks may need to return 12% or more to offset bond returns that are likely to be less than half that. Alternatively, some investors are concentrating their portfolios more heavily in stocks because of the low bond yields.

Nonetheless, many investors assume that stock market returns over this decade will be inside the range of 8% to 12%. As Figure 2 reflects, the annualized total return from the stock market over rolling 10-year periods has been inside the 8% to 12% range less than one-quarter of the time!

Source: “2011 Ibbotson Stocks, Bonds, Bills and Inflation Classic Yearbook” (Morningstar, 2011) and Crestmont Research.
SPECIAL OFFER: Get AAII membership FREE for 30 days!
Get full access to AAII.com, including our market-beating Model Stock Portfolio, currently outperforming the S&P 500 by 2-to-1. Plus 60 stock screens based on the winning strategies of legendary investors like Warren Start your trial now and get immediate access to our market-beating Model Stock Portfolio (beating the S&P 500 2-to-1) plus 60 stock screens based on the strategies of legendary investors like Warren Buffett and Benjamin Graham. PLUS get unbiased investor education with our award-winning AAII Journal, our comprehensive ETF Guide and more – FREE for 30 days

Just over one-third of the time (35%), annualized 10-year returns exceeded 12%. The remainder of the 10-year periods, 43%, delivered annualized returns that were less than 8%.

In Figure 2, the columns list the respective 10-year periods. The periods are not random; rather, they run in series that reflect the fluctuating trend in the price-earnings ratio as driven by the inflation rate. In the rightmost column, reflecting periods with returns in excess of 12%, are all periods when the price-earnings ratio started at less than 15. For the leftmost column, with returns below 8%, the starting price-earnings ratio was generally above 15. With the price-earnings ratio currently near 20, an argument could be made that the next 10-year period ultimately will find its place in the left column.

Driver, Take Me To…

There are two major drivers of stock market returns. The first is the growth rate in earnings. The second is the overall level of valuation in the stock market, as measured by the price-earnings ratio. The price-earnings ratio has two significant effects: The price-earnings ratio drives dividend yield (higher valuations make cash dividends reflect lower-yield percentages), and higher price-earnings ratios are vulnerable to decline just as low price-earnings ratios offer the opportunity of multiplied returns from rising valuations.

Although corporate profits ride a wild business cycle, ultimately high and low margins return to normal. As a result, the long-term growth rate of earnings is similar to that of revenues, since profit margins are determined by dividing profits by revenues. Further, aggregate revenue growth reflects economic growth. Therefore, earnings growth is driven by economic growth.

One of the major uncertainties for the future is economic growth, and thus earnings growth. It is an uncertainty that weighs heavily on the stock market.

The second major uncertainty relates to the second major driver of stock market returns: changes in the price-earnings ratio of the overall stock market. The most significant driver of market price-earnings ratio over time is the level and trend in the inflation rate. High inflation brings high interest rates and lower price-earnings ratios. Deflation brings declining earnings in nominal (non-inflation adjusted) terms and thus lower price-earnings ratios. Conversely, high price-earnings ratios are driven by low and stable inflation, reflecting positive financial conditions and greater certainty. But today’s relatively low inflation confronts the tightly bound tectonic plates of deflation and inflation, which could suddenly surge in either direction. Economic growth and inflation are also items where the conventional confronts the unconventional.

Pause for a moment to take a pop quiz; let’s use it to reveal a few insights.

Over the past century in the United States, real economic growth before inflation has averaged near 3% per year. Over the decades of the 1970s, 1980s and 1990s, the compounded average annual growth rate was 3.2%, 3.0% and 3.2%, respectively. So during the decade of the 2000s (2000–2009), when consumers were loading up their credit cards, homeowners were said to be using home equity like an ATM, unemployment averaged 5.5% and fell below 4% at times, and debt was being added to debt, what was the compounded economic annual growth rate before inflation rounded to the nearest percent?

A) 4%
B) 3%
C) 2%

For many, the first choice, 4%, is the most logical response. It reflects the perception that much of the consumption and leverage of the 2000s artificially accelerated economic growth. People who choose 4% expect that the factors present in the 2000s boosted economic growth above the historical and recent average growth rates.

Following such a strong period of economic growth, most people answering “A” expect a period of below-average growth over this decade to make up for the excesses of the prior decade. They expect that periods during which growth was fueled by debt will be followed by offsetting moderation as the vestiges of leverage and excess consumption are addressed.

The second choice, 3%, is the contrarian response. It reflects a belief that the last 10 years were not different from the previous decades. Though some of the factors in play in the 2000s may have impacted economic growth, people who choose 3% either don’t believe that those factors had much effect, or presume that there may have been similarly unique factors during prior decades. Nonetheless, economic growth of 3% has endured for more than 100 years and has been very consistent in recent decades. Some people in this group believe that 3% is likely for this decade, while others have begun to adopt the notion of a “new normal” of slowing growth due to recent trends in demographics, government policy, taxes, etc.

The third choice, 2%, is the correct response, despite being the one least often selected. Many investors are surprised that the decade of the 2000s experienced compounded annual growth of only 1.8%. Some economists say that it was a decade sandwiched by two recessions, while others blame it on the severe recession of 2008 and the related financial crisis. Yet excluding the recession of 2008 from the decade, the growth rate for the first eight years of the 2000s was still only 2.6%. Further, cumulative economic growth throughout the decade of the 2000s did not exceed 2.7%. It would have required an unusual surge—near 4.5% annually—in the final two years for the full decade to reach the historical average annual growth rate of near 3%.

This sets the stage for a dilemma. Will this decade restore the long-term average by growing at 4%, thereby defying the predominant belief in a slow-growth decade? Was the prior decade an anomaly, with future economic growth simply returning again to its long-term trend of 3%? Did something change 10 years ago, and has economic growth downshifted to a level near 2%? Or as some might contend, could the rate be even lower due to the economic, financial, and/or policy headwinds in front of us? All three scenarios are plausible, which makes economic growth a major uncertainty. The answer to the dilemma has very significant implications for stock market returns over the next decade and longer.

Projected Inflation

But the pop quiz isn’t over: What’s your outlook for inflation? Figure 3 presents a picture of the inflation rate for more than a century.

The roller-coaster ride reflects an undulating cycle that does not ride a consistent wave of time. There’s no right answer for this question either … at least not for another 10 years. But there are three plausible scenarios. Does your outlook include a trend of deflation (the purple line in the graph), the calm of price stability (the green line), or a trend of rising inflation (the red line)? If the answer is not clear—and it is not for most people—then certainly you now appreciate the second major uncertainty: What will the rate of inflation be in the future?


Based upon current levels of valuation, an argument could be made that the stock market is likely to deliver modest or minimal returns. Nonetheless, investors should not avoid stocks. They should diversify their portfolios and enhance returns from stock market investments.

To hedge inflation risk in portfolios, investors can complement diversification with TIPS (Treasury Inflation-Protected Securities), inflation-sensitive commodity investments, and defensive stocks that offer higher dividend yields. To address deflation risk, investors will find that zero-coupon Treasury bonds provide near-guaranteed returns if held to maturity, yet in an environment with significant deflation those bonds can double or triple in value in the near term. If inflation remains under control, investors can sell call options on stock portfolios to enhance returns while still participating in capital gains.

These are among the many ways that investors can realize higher returns when the stock market treads through periods of below-average returns. Using a boating analogy, investors should pull out the oars and actively row their portfolios toward success, while waiting for the next long-term bull market to more passively sail portfolios into the sunset.

Ed Easterling is the author “Probable Outcomes: Secular Stock Market Insights” (Cypress House Press, 2011). Further, he is president of Crestmont Research and a senior fellow with the Alternative Investment Center at Southern Methodist University’s Cox School of Business.


H from NC posted over 4 years ago:

Great analysis - provides some valuable insights!

Edward from PA posted over 4 years ago:

Good article. This agrees with John Hussman's analysis of average annual returns of less than 5% for the next decade. See his weekly comments on the Hussman Funds web site

David from MI posted over 4 years ago:

Excellent article. I was surprised that actual economic growth in the last decade was 2%; I assumed it was 3% even with two recessions during the ten year period.

James from OH posted over 4 years ago:

I agree with Mr. Easterling’s analysis. Averages can be misleading. For example, let’s say yesterday the temperature was 80°. I was too hot. Today, the temperature is 60°. Now, I’m too cold. However, the average of 70 was just right.

In order to portray the variation in the average annual returns (AAR) that an investor would have obtained, Mr. Easterling categorized all 10-year returns since 1926 as either less than 8%, 8%-12%, or more than 12%. While this is effective, I feel that it understates the magnitude of the variation in the reader’s mind.

I came to this conclusion by looking at the “moving” AAR for each of 10-year, 20-year, 30-year, and 40-year time periods. I did so using Robert Shiller’s data which can be obtained at www.econ.yale.edu/~shiller/data.htm. This is monthly data for the S&P 500 going back to 1871. It includes dividends. The AARs are “moving” in that for each month in the 140-year period, I calculated the AAR over each of the next 10, 20, 30, and 40 years.

The average AARs for each of the four time periods starting in 1926 are fairly similar. They are 10.5%, 11.2%, 11.2%, and 10.8%, respectively for the 10, 20, 30, and 40 year periods. This is consistent with the roughly 10% AAR that Mr. Easterling quoted and other studies have found for the stock market over the long term.

However, the minimum and maximum AARs vary quite a bit from the 10% average. The minimums were -4.0%, 2.0%, 7.6%, and 7.9%. The maximums were 21.2%, 17.9%, 14.3%, and 13.2%.

These results are more vividly portrayed in the following graph. (I could not figure out how to post the graph in this commments section of the AAII web site. If you want a copy, send me an email. The graph portrays a fairly consistent up and down cycle for each of the four time periods.)

The conclusion here is fairly obvious. The AAR you can reasonable expect over the long run clearly depends on (1) When you make the investment and (2) How long it is invested.

This, of course begs the question, “Where are we now?” If one looks at the 10-year, 20-year, 30-year, and 40-years lines in the graph, it appears that peaks in AAR occur roughly every 35 years. As the last bottom occurred about 15 years ago, that suggests that we are somewhere near a peak right now. Hence, an investor would likely be too optimistic to assume an AAR in the future near the long term average of 10%. An AAR from -4% to 8%, depending on his investment horizon, is likely to be more in line.

Jim Grant
Solon, Ohio

Lance from NC posted over 4 years ago:

The "new normal" as PIMCO exec's call it is for lower growth with lower long term returns based upon where we are and the analysis. We have a mature economy and with maturity in markets for companies comes lower long term growth unless we find new markets to feed growth, unfortunately.

Dave from WA posted over 4 years ago:

Great article and follow up post by James!

George Pettiford from IL posted over 2 years ago:

The future is unpredictable; there are always BLACK SWAMS ever more present. Just ask the banks or the "quants"

David Van Knapp from NY posted 5 months ago:

Do the returns included reinvestment of dividends or just the collection of them? My impression is that the returns as shown do not include the effects of reinvesting dividends.

It is an important distinction, because dividends reinvested during "bad times" (low prices) have more impact on true total returns (i.e., returns including the impact of reinvesting dividends) than they do when stock prices are relatively high.

You need to log in as a registered AAII user before commenting.
Create an account

Log In