Case Closed: Stocks Work
by Gene Epstein
Investing for retirement -- If past is prologue, returns over the next five and 10 years will be better than average -- so don't give up yet.
This year marks the 30th anniversary of a famous Business Week cover story "The Death of Equities." Along with those scary words, the magazine's Aug. 13, 1979, cover read, "How inflation is destroying the stock market." But starting around that time, investors could have beaten inflation quite handily by snapping up stocks and holding them for five or 10 years.
Buying the stock market at the close of 1979 would have yielded, after inflation, an average annual return of 7.3% over the next five years. An even higher five-year return of 9.47% could have been captured by going long at the end of 1978. The 10-year performance would have been healthier still, yielding 9.52% or 10.75%, depending on whether the investor bought at the close of '78 or '79.
With the stock market in the throes of yet another near-death experience, another rebirth could be in the offing. Five- and 10-year returns on stocks through year-end 2008 have run negative, and would have looked even worse at the lows of last week. But based on the historical record, performances like these bode well for the next five to 10 years.
The historical record shows that for 20- and 30- year periods, inflation-adjusted returns on stocks have never been negative. Over the 137 years from 1871 through 2008, returns after inflation for 20- and 30-year intervals have been consistently positive. Median returns over the 20-year intervals have been 6.85%, and for 30-year intervals, 6.23%.
With this consistently strong performance over long periods, it stands to reason that below-par returns over five- and 10-year intervals would tend to be followed by much better results over the subsequent five- and 10-year intervals. And in fact, the historical record shows that, following below-average returns over five and 10 years, subsequent periods of similar length do tend to perform better than average.
An investor whose retirement is drawing near might take heed: Investing in stocks today could help produce the cash you will need five or 10 years down the road.
Those who plan to retire in less than 10 years would benefit if the historical trends hold true. Positive returns over the next 20 or 30 years would only make retirement more of a breeze.
Critics of stocks as vehicles for retirement often rig their case by assuming that investors entered and exited with the worst possible timing, buying at peaks and liquidating at bottoms.
But diversification over time -- buying and selling periodically, rather than all at once -- can be quite effective. Most investors would be foolish to liquidate all their stock holdings on the day their retirement begins, unless they feel endowed with timing skills that few can claim. If they plan to live 20 years past their retirement, they might plan to hold on to at least part of their holdings for 15 to 20 years.
And of course, retirement accounts are set up in such a way that buying can occur in installments over many years. The acquisition of stocks can therefore be diversified over time, along with the process of liquidation.
From this perspective, useful insights can be gleaned from the exhaustive record originally pieced together by Wharton School finance professor Jeremy Siegel for his best-selling book, Stocks for the Long Run, now in its fourth edition.
Siegel has amassed data on rolling five-year periods dating back to 1871 (1871-1876, 1872-1877 and so on). He has similar data on rolling 10-, 20- and 30-year periods.
Why begin with 1871? Prof. Siegel can also provide data going back to 1802, but prior to 1871, the quality of the data isn't particularly reliable, and data over the past 137 years are more than sufficient to reveal the long-term performance of stocks as an asset class.
The data can track all failed stocks into bankruptcy, so there is no "survivors' bias," a common flaw in historical analysis. And Siegel adds that, even in the 1800s, the U.S. stock market featured a fair range of different industries, roughly similar to more recent eras.
Siegel has analyzed the data in terms of "total returns" after inflation. All publicly traded stocks are bought on a capitalization-weighted basis, with all dividends reinvested. Average annual returns benefit from the magic of compounding. Thus, for example, $1 invested at 6.26% over 30 years becomes an inflation-adjusted $6.13 with compounding.
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. Perhaps a tad unrealistically, management fees aren't factored in, either. But in the era of index funds and exchange-traded funds, such fees are lower than ever. Some ETFs charge as little as seven one-hundredths of a percent.
Jeremy Schwartz, research director of WisdomTree Asset Management -- a firm with which Siegel is affiliated -- updated Siegel's figures at Barron's request. We asked him to line up the worst-performing quartile of 10-year stretches since 1971 and then see how the following 10 years performed in each case. That meant examining about 30 intervals of poor performance.
The result: In each case -- without exception -- the subsequent 10-year periods performed better and ran positive. The median performance for each was 8.17%, 1.33 percentage points higher than the median for all 10-year intervals.
Schwartz performed the same exercise for the worst quartile of five-year returns. Here the finding was that, in 25 out of the 31 cases, the subsequent five-year periods performed better and ran positive. The median performance for all these cases was 9.47%, 2.50 percentage points higher than the median for all five-year intervals.
Prof. Siegel also compares long-term equity performance with returns in U.S. Treasury bonds -- an apt comparison for risk-averse investors seeking reliable income in retirement. Assuming buy-and-hold strategies in Treasuries over 20- and 30-year intervals, how often did the inflation-adjusted income and possible capital gains from bonds prove superior to the returns of stocks?
Answer: Through 2008, stocks have always done better than Treasury bonds over 30-year periods. And over 20 years, stocks bested Treasuries in all but a little over 5% of the cases.
Despite the bear market of 2008, long-term returns through year end were fairly good, running 5.17% annually for the previous 20 years and 6.6% for the previous 30. But what if the investor had the bad luck to liquidate at of the close of February '09? Add these two disastrous months to the 20- and 30-year holding periods, and returns would have been 4.09% and 5.86%, respectively.
Why do stocks tend to do better over the long run than either bonds or inflation? Mainly because their returns are driven by rising profits -- which in turn are driven by real growth in the U.S. economy. That's why stocks can be indispensable for retirement planning.
Copyrighted, Dow Jones & Company, Inc. All Rights Reserved.
Investing for retirement -- If past is prologue, returns over the next five and 10 years will be better than average -- so don't give up yet.
This year marks the 30th anniversary of a famous Business Week cover story "The Death of Equities." Along with those scary words, the magazine's Aug. 13, 1979, cover read, "How inflation is destroying the stock market." But starting around that time, investors could have beaten inflation quite handily by snapping up stocks and holding them for five or 10 years.
Buying the stock market at the close of 1979 would have yielded, after inflation, an average annual return of 7.3% over the next five years. An even higher five-year return of 9.47% could have been captured by going long at the end of 1978. The 10-year performance would have been healthier still, yielding 9.52% or 10.75%, depending on whether the investor bought at the close of '78 or '79.
With the stock market in the throes of yet another near-death experience, another rebirth could be in the offing. Five- and 10-year returns on stocks through year-end 2008 have run negative, and would have looked even worse at the lows of last week. But based on the historical record, performances like these bode well for the next five to 10 years.
The historical record shows that for 20- and 30- year periods, inflation-adjusted returns on stocks have never been negative. Over the 137 years from 1871 through 2008, returns after inflation for 20- and 30-year intervals have been consistently positive. Median returns over the 20-year intervals have been 6.85%, and for 30-year intervals, 6.23%.
With this consistently strong performance over long periods, it stands to reason that below-par returns over five- and 10-year intervals would tend to be followed by much better results over the subsequent five- and 10-year intervals. And in fact, the historical record shows that, following below-average returns over five and 10 years, subsequent periods of similar length do tend to perform better than average.
An investor whose retirement is drawing near might take heed: Investing in stocks today could help produce the cash you will need five or 10 years down the road.
Those who plan to retire in less than 10 years would benefit if the historical trends hold true. Positive returns over the next 20 or 30 years would only make retirement more of a breeze.
Critics of stocks as vehicles for retirement often rig their case by assuming that investors entered and exited with the worst possible timing, buying at peaks and liquidating at bottoms.
But diversification over time -- buying and selling periodically, rather than all at once -- can be quite effective. Most investors would be foolish to liquidate all their stock holdings on the day their retirement begins, unless they feel endowed with timing skills that few can claim. If they plan to live 20 years past their retirement, they might plan to hold on to at least part of their holdings for 15 to 20 years.
And of course, retirement accounts are set up in such a way that buying can occur in installments over many years. The acquisition of stocks can therefore be diversified over time, along with the process of liquidation.
From this perspective, useful insights can be gleaned from the exhaustive record originally pieced together by Wharton School finance professor Jeremy Siegel for his best-selling book, Stocks for the Long Run, now in its fourth edition.
Siegel has amassed data on rolling five-year periods dating back to 1871 (1871-1876, 1872-1877 and so on). He has similar data on rolling 10-, 20- and 30-year periods.
Why begin with 1871? Prof. Siegel can also provide data going back to 1802, but prior to 1871, the quality of the data isn't particularly reliable, and data over the past 137 years are more than sufficient to reveal the long-term performance of stocks as an asset class.
The data can track all failed stocks into bankruptcy, so there is no "survivors' bias," a common flaw in historical analysis. And Siegel adds that, even in the 1800s, the U.S. stock market featured a fair range of different industries, roughly similar to more recent eras.
Siegel has analyzed the data in terms of "total returns" after inflation. All publicly traded stocks are bought on a capitalization-weighted basis, with all dividends reinvested. Average annual returns benefit from the magic of compounding. Thus, for example, $1 invested at 6.26% over 30 years becomes an inflation-adjusted $6.13 with compounding.
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. Perhaps a tad unrealistically, management fees aren't factored in, either. But in the era of index funds and exchange-traded funds, such fees are lower than ever. Some ETFs charge as little as seven one-hundredths of a percent.
Jeremy Schwartz, research director of WisdomTree Asset Management -- a firm with which Siegel is affiliated -- updated Siegel's figures at Barron's request. We asked him to line up the worst-performing quartile of 10-year stretches since 1971 and then see how the following 10 years performed in each case. That meant examining about 30 intervals of poor performance.
The result: In each case -- without exception -- the subsequent 10-year periods performed better and ran positive. The median performance for each was 8.17%, 1.33 percentage points higher than the median for all 10-year intervals.
Schwartz performed the same exercise for the worst quartile of five-year returns. Here the finding was that, in 25 out of the 31 cases, the subsequent five-year periods performed better and ran positive. The median performance for all these cases was 9.47%, 2.50 percentage points higher than the median for all five-year intervals.
Prof. Siegel also compares long-term equity performance with returns in U.S. Treasury bonds -- an apt comparison for risk-averse investors seeking reliable income in retirement. Assuming buy-and-hold strategies in Treasuries over 20- and 30-year intervals, how often did the inflation-adjusted income and possible capital gains from bonds prove superior to the returns of stocks?
Answer: Through 2008, stocks have always done better than Treasury bonds over 30-year periods. And over 20 years, stocks bested Treasuries in all but a little over 5% of the cases.
Despite the bear market of 2008, long-term returns through year end were fairly good, running 5.17% annually for the previous 20 years and 6.6% for the previous 30. But what if the investor had the bad luck to liquidate at of the close of February '09? Add these two disastrous months to the 20- and 30-year holding periods, and returns would have been 4.09% and 5.86%, respectively.
Why do stocks tend to do better over the long run than either bonds or inflation? Mainly because their returns are driven by rising profits -- which in turn are driven by real growth in the U.S. economy. That's why stocks can be indispensable for retirement planning.
Copyrighted, Dow Jones & Company, Inc. All Rights Reserved.
Comments