U.S. stocks returned 12.5% last year, including reinvested dividends—well above the long-term average of 8.9%, going back to 1871. Five-year average annual returns on the stock market through 2014 ran even higher, at 15.8%.
 
Illustration: Stuart Goldenberg for Barron's
 
 
Over the long term, there is a pronounced tendency for periods of better-than-average returns on equities to be followed by stretches of worse-than-average, and vice versa. Accordingly, the above-par performance of the past several years suggests that, over the next several years, performance will run somewhat below-par, but still in the black.
 
Barron’s has weighed in on the historical long-term return data before, beginning in March 2009, when five-year returns were in negative territory. Accordingly, we predicted that the market’s recent near-death experience would most likely be followed by a resurgent rebirth, (“Case Closed: Stocks Work,” March 9, 2009). By dint of uncanny prescience—or maybe dumb luck—the publication date of the article coincided with the very day the market averages hit cyclical lows.
 
But if the last five years were good, they weren’t great, and that bodes well for the future.
 
THE MAIN SOURCE of these findings is Wharton School finance professor Jeremy Siegel, author of the aptly titled best seller, Stocks for the Long Run. With the help of one of his former students, Jeremy Schwartz—research director of WisdomTree Investments  (ticker: WETF), a firm with which Siegel is associated—the Wharton professor has amassed figures on equity returns since 1871, the earliest year for which reliable data are available. Starting at the close of calendar year 1870, Schwartz compiled returns for rolling five-year periods: 1871-1876, 1872-1877, and so on, culminating in 2009-2014—139 five-year stretches in all.
                
cat
 
As the table on this page shows, the median average annual performance over these 139 periods is 9.5%. And while the past five years is in the top quartile of returns, it is only just barely. The best five-year performance was 27%, during the Roaring Twenties, from 1923 through 1928. The worst five-year interval—negative 15.6% a year—ran from 1927 through 1932, coinciding with the Great Crash of ’29.
 
Comments Siegel: “Although the returns in the last five years have been extraordinary, these returns came on the heels of the deepest bear market in 75 years that resulted in stocks trading at extremely undervalued levels.”
 
Noting that the recent five years belongs in the top quartile of returns, Schwartz did a simple follow-up analysis. After separating out the 35 intervals in the top quartile, he examined the five years that immediately followed each of these periods. Over these spans, he found that the median return was 9%.
 
Backing out a two percentage-point contribution from dividends, we get a median expectation of 7% annual gains in stock prices over the next five years.
 
Further bolstering the case for an upbeat outlook on stock prices: 10- and 15-year intervals are still below average to subpar, which implies above-par performance over the next 10 to 15 years. As the table shows, 10-year returns through 2014 ran at an annual rate of 8.5%, slightly below the median returns on all 10-year intervals of 8.6%. And 15-year returns of 5.2% are in the bottom quartile of all 15-year returns. As Schwartz notes, median 15-year returns following all cases in the bottom quintile ran 10%.
 
THIS 144-YEAR RECORD makes a compelling case for stocks over the long run. As the table shows, just 16 out of 139 five-year intervals ran in the red; even after adjusting for inflation, there were only 25 periods of loss out of 139. For 15-year intervals, there are no periods of loss, and just three after inflation adjustment. For 20- and 30-year periods, all cases are profitable, whether or not you inflation-adjust.
 
Failed stocks are tracked into bankruptcy, so there is no “survivor’s bias,” a common flaw in such historical analysis. And Siegel notes that, certainly by the late-1800s, the U.S. stock market featured a fair range of different industries, roughly similar to more recent eras. All publicly traded stocks are bought on a capitalization-weighted basis, with all dividends reinvested. For any given holding period from year-end close to year-end close, no taxes are assumed—not unrealistic, given the advent of tax-deferred accounts.
 
Less realistically, no investment fees are factored in, but in the era of index funds and ETFs, such fees can be quite small.
 
The underlying assumption from this data is that past is prologue, that the past 144 years of equity performance will be about the same as the next 144. Assuming the same 8.9% (actually, 8.85%) average annual return for the next 144 years, $10,000 invested in the stock market today will turn into more than $2 billion—probably still real money even then.